feat: Optimal P source provenance and unified console/PNG label format

[nerdCopter]

Summary

Builds on the Reliable/Unreliable label fix from #150 (fixes #149) by adding source provenance, Setpoint Authority classification, and fully aligning console and PNG label layout.

Optimal P source provenance

• Add source_events and source_files fields to OptimalPAnalysis, populated from AircraftProfile.file_count and AxisProfile.event_count
• Td source: line: File Group when data pooled across ≥2 files; Single File for single-file runs
• Preamble (Td target: physics-derived from throttle-punch events in log group|file.) printed once per aircraft in console; removed as a separate PNG legend line (redundant with Td source:)

Setpoint Authority classification

• Replace binary [LOW AUTHORITY] flag with always-visible Setpoint Authority: line (console + PNG)
• New SetpointAuthority enum (Low/Moderate/High) and compute_setpoint_authority() in calc_step_response.rs
• Uses mean of per-window max setpoints (not max) — a single outlier input no longer masks a predominantly hover-style flight
• Thresholds: LOW < 100 dps, MODERATE 100–250 dps, HIGH > 250 dps (constants in constants.rs)
• All P:D recommendations always shown regardless of authority level; pilot reads the authority line and decides
• Orange for LOW, neutral for MODERATE/HIGH

Label line order — console and PNG (identical content)

Roll: Actual Peak=X.XXX → Assessment

  Current P:D=X.XX (Actual Peak=X.XX, Assessment)
  ─────────────────────────────
  Setpoint Authority: LOW|MODERATE|HIGH (mean=Xdps ⊢≥100dps)
  Recommendation (conservative): P:D=X.XX (D≈X)
  Recommendation (moderate): P:D=X.XX (D≈X)
  Recommendation (aggressive): P:D=X.XX (D≈X)   [significant overshoot only]
  Recommendation (none): No obvious tuning adjustments needed
  ─────────────────────────────
  Optimal P (Experimental, log-derived)
    Td: X.Xms (target: X.X±X.Xms, windows=N)
    Td source: File Group|Single File — N flights, M throttle-punches
    Noise: LOW|MODERATE|HIGH
    Deviation: ±X.X% (ZONE)
    Current P=N
    Recommendation (Conservative|Decrease): P≈N (±N)
    Reliable|Unreliable: Consistency=N% (⊢≥70%), CV=N.N% (⊢≤40%)

PNG legend refinements

• Curve labels: Step-Response Avg (Smoothed Peak: X.XX, Td: Xms); < N deg/s (Smoothed Peak: …) / ≥ N deg/s (Smoothed Peak: …)
• ───────────────────── separator between curve entries and P:D section (consistent with Optimal P divider)
• 2-space indent on all lines below Current P:D=…

Console/PNG alignment

• Peak= → Actual Peak= (matches PNG); recommendation format uses () not →; (P=N) suffix removed
• Blank line after axis header in console

Test plan

• Build passes, clippy clean, cargo fmt no-op
• Single-file run: source line shows Single File — 1 flight, N throttle-punches
• Directory/group run: source line shows File Group — N flights, M throttle-punches
• Hover-test log: Setpoint Authority: LOW (mean=Xdps ⊢≥100dps) shown in orange; all recommendations still visible
• Normal flight: Setpoint Authority: MODERATE or HIGH in neutral color
• PNG legend matches console content for all recommendation arms
• Smoothed Peak on curve labels; Actual Peak on Current P:D assessment line
• ───────────────────── separator visible between curve entries and P:D block

🤖 Generated with Claude Code

[coderabbitai]

Important

Review skipped

Auto incremental reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: ffb396a8-d3bb-44d6-bbac-d7d6ca584693

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

📝 Walkthrough

Walkthrough

This PR extends the optimal P analysis with provenance tracking. OptimalPAnalysis and AircraftProfile gain new fields to track source event counts and contributing file counts. These fields are populated during profiling and surfaced in console and plot legend outputs alongside analysis recommendations.

Changes

Provenance Tracking and Reporting

Layer / File(s)	Summary
Data model extensions src/data_analysis/optimal_p_estimation.rs, src/data_analysis/torque_inertia_profiler.rs	OptimalPAnalysis struct adds source_events and source_files fields; AircraftProfile adds file_count field to track the number of contributing flight files.
Field initialization and population src/data_analysis/optimal_p_estimation.rs, src/data_analysis/torque_inertia_profiler.rs, src/main.rs	Constructors initialize new fields to 0; profile_aircraft_group sets file_count from the count of successfully profiled files; optimal P analysis success path populates source_events and source_files from aircraft profile axis metrics before storing.
Console and plot output formatting src/data_analysis/optimal_p_estimation.rs, src/plot_functions/plot_step_response.rs	format_console_output is rewritten to include a structured "Source" line and reorganized report layout; plot legend sections are updated to display source (group vs single), event counts, and revised Td/noise/tolerance formatting alongside recommendations and reliability status.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Possibly related PRs

• nerdCopter/BlackBox_CSV_Render#150: Both PRs directly modify Td "reliability"/consistency formatting in src/data_analysis/optimal_p_estimation.rs's format_console_output and the Optimal P legend reliability output in src/plot_functions/plot_step_response.rs, so the changes overlap strongly at the same code paths and user-facing strings.
• nerdCopter/BlackBox_CSV_Render#145: The main PR refines the existing "optimal P" implementation by updating OptimalPAnalysis/format_console_output, extending AircraftProfile with file_count, and reworking the plot_step_response optimal-P legend output.
• nerdCopter/BlackBox_CSV_Render#143: Both PRs update src/plot_functions/plot_step_response.rs's "recommendation" legend/console text and D recommendation formatting, so the changes directly overlap at the step-response recommendation output level.

Suggested labels

enhancement

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'feat: Optimal P source provenance and unified console/PNG label format' accurately summarizes the two main changes in this PR: adding source provenance tracking and unifying label formatting between console and PNG outputs.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/optimal-p-source-info

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@src/data_analysis/optimal_p_estimation.rs`:
- Around line 509-518: The console label generation using output.push_str(...)
emits an axis-prefixed Td line but omits the "Optimal P (Experimental,
log-derived)" header, causing structural mismatch with the PNG block; update the
formatter that builds the label (the code using output.push_str and the symbols
axis_name, self.td_stats, self.td_target_ms, self.td_tolerance_ms,
self.td_stats.num_samples, self.noise_level.name()) so it first emits the
"Optimal P (Experimental, log-derived)" header (matching the PNG block) and then
emits per-axis Td/Noise lines in the same order/structure as the PNG output,
ensuring verbatim parity between console and PNG content.

In `@src/main.rs`:
- Around line 515-517: The profile.file_count is being set to files_profiled
which counts any parsable file, overstating flights; change this to count only
files that actually contributed throttle-punch events (e.g., introduce or use an
existing counter like files_with_contribs or compute from the collection of
per-file event counts) and assign that value to profile.file_count instead of
files_profiled; ensure the increment of that counter happens only when a file
produces a non-zero number of throttle-punch events (update the logic that
currently increments files_profiled to increment the new counter only on
contribution).

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: d7a850b5-5107-4625-b311-7e2e5f163b78

📥 Commits

Reviewing files that changed from the base of the PR and between f1d13b7 and 78cd6df.

📒 Files selected for processing (4)

• src/data_analysis/optimal_p_estimation.rs
• src/data_analysis/torque_inertia_profiler.rs
• src/main.rs
• src/plot_functions/plot_step_response.rs

📜 Review details

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: Test and Lint

🧰 Additional context used

📓 Path-based instructions (1)

src/**/*.rs

📄 CodeRabbit inference engine (AGENTS.md)

src/**/*.rs: All constants go in src/constants.rs — no hardcoded magic numbers in function code
Axis indices (0=Roll, 1=Pitch, 2=Yaw) must use the src/axis_names.rs module with AXIS_COUNT, AXIS_NAMES, and axis_name() function instead of creating redundant axis index constants
ALWAYS use Firmware revision header to detect firmware type, NEVER use Firmware type
Firmware revision patterns: Betaflight uses Betaflight 4.x.x or Betaflight YYYY.MM, Emuflight uses EmuFlight 0.x.x, INAV uses INAV x.x.x

Files:

• src/data_analysis/torque_inertia_profiler.rs
• src/plot_functions/plot_step_response.rs
• src/data_analysis/optimal_p_estimation.rs
• src/main.rs

🔇 Additional comments (3)

src/data_analysis/torque_inertia_profiler.rs (1)

116-117: LGTM!

Also applies to: 130-130

src/main.rs (1)

1252-1255: LGTM!

src/plot_functions/plot_step_response.rs (1)

590-590: LGTM!

Also applies to: 594-604, 609-622, 627-649, 651-666, 671-703, 706-740