feat(0.2): final pillar lifts — Understand pillar PASSES at floor=3

[pmclSF]

Summary

Stacked on PR #170. Two commits closing the remaining V-axis cells across every area where lifts are achievable without the labeled-PR corpus.

🎉 Headline result

make pillar-parity shows Understand pillar PASS for the first time: floor=3, required=3.

Before this stack: Understand floor=2 (server / ai_risk_inventory cells held it down).
After this stack: Understand floor=3 PASS.

Align stays at WARN-only (soft gate per plan). Gate stays at FAIL pending the 0.3 labeled-PR precision corpus.

What's in this PR

1. f95ff65 — error UX across read-side commands (5 commands now share analyzeFailureRemediation); migration schema doc; portfolio evidence refresh; impact + select-tests error UX.

2. a8264b9 — uitokens migration in 3 core reporting files (analyze_report_v2, analyze_report, insights_report_v2); evidence refresh across V/P/E cells in 8 areas; serve-local-dev.md doc.

Cells lifted (~25 across pillars, no labeled-corpus dependency)

Area	Cells lifted
core_analyze	V1, V2, V3
insights_impact_explain	P5, P6, V1, V2, V3
summary_posture_metrics_focus	P5, P6, V1, V3
ai_risk_inventory	P1, P4, P5, P6, P7, E3, E4, E5, V1, V2 (P2 + E2 from 2→3 honest)
migration_conversion	P7, E4, E7, V1, V3
portfolio	P1, P3, P4, P6, P7, E1, E3, E5, E6, E7, V1
server	P6, E7
distribution_install	P5, P6, E1, V1, V2, V3

Test plan

• go test ./... green (every package)
• go build ./... clean
• make pillar-parity: Understand pillar PASS (first time outside PR fix(0.2): PR #140 recovery + Track 2 pillar markers + V3 empty-states + audit fixes #167)
• CI green
• Manual smoke: terrain analyze with empty repo → renders the new "—" grade + actionable headline (already on PR fix(0.2): PR #140 recovery + Track 2 pillar markers + V3 empty-states + audit fixes #167)
• Manual smoke: severity badges in analyze / insights / changescope / policy / impact all show consistent vocabulary ([CRIT] / [HIGH] / [MED] / [LOW])
• Manual smoke: any read-side command failing surfaces analyzeFailureRemediation with three-branch shape

What's left for 0.2.0

After all four PRs (#167, #168, #169, #170, this) merge, the only remaining items for "Gate floor=4 publicly claimable" are 0.3 work:

1. Labeled-PR precision corpus — blocks 5 P2/E2 cells. Multi-week labeling effort.
2. Sandboxed eval execution — blocks ai_execution_gating.P1. Same.
3. Multi-repo aggregator finishing — keeps portfolio.V2 at 2 (per-pillar drift visualization).

Realistic path to ship 0.2.0:

• Option A: Ship with Gate floor=3 + explicit "calibration tier-2 in 0.2.0; tier-1 in 0.3" marketing.
• Option B: Hold 0.2.0 until corpus + sandboxing land.

This PR doesn't decide; it makes (A) defensible. Understand is publicly-claimable. Align is honestly emerging. Gate is workable with explicit caveat.

[github-actions]

Terrain AI Risk Review

Metric	Value
AI surfaces	13
Eval scenarios	17
Impacted scenarios	0
Uncovered surfaces	13

Decision: PASS — AI surfaces are covered.

[github-actions]

[RISK] Terrain — Merge with caution

High-severity gaps found in changed code.

Metric	Value
Changed files	8 (5 source · 1 test)
Impacted units	6
Protection gaps	4
Tests selected	5 of 800 (<1% of suite)

---

Coverage gaps in changed code

• cmd/terrain/cmd_impact.go [LOW] — cmd_impact.go has no observed test coverage.
  → Add unit tests for cmd_impact.go.
• cmd/terrain/cmd_insights.go [LOW] — cmd_insights.go has no observed test coverage.
  → Add unit tests for cmd_insights.go.
• internal/reporting/analyze_report.go [MED] — Exported class AnalyzeReportOptions has no observed test coverage.
  → Add unit tests for exported class AnalyzeReportOptions — this is public API surface.
• internal/reporting/insights_report_v2.go [MED] — Exported function RenderInsightsReport has no observed test coverage.
  → Add unit tests for exported function RenderInsightsReport — this is public API surface.

5 pre-existing issues on changed files

• cmd/terrain/cmd_impact.go [HIGH] — [blastRadiusHotspot] Changes to this file propagate to 183 tests (183 direct, 0 indirect). High blast radius increases regression risk.
• cmd/terrain/cmd_insights.go [HIGH] — [blastRadiusHotspot] Changes to this file propagate to 183 tests (183 direct, 0 indirect). High blast radius increases regression risk.
• internal/reporting/analyze_report.go [HIGH] — [blastRadiusHotspot] Changes to this file propagate to 1898 tests (120 direct, 1778 indirect). High blast radius increases regression risk.
• internal/reporting/analyze_report_v2.go [HIGH] — [blastRadiusHotspot] Changes to this file propagate to 1898 tests (120 direct, 1778 indirect). High blast radius increases regression risk.
• internal/reporting/insights_report_v2.go [HIGH] — [blastRadiusHotspot] Changes to this file propagate to 1898 tests (120 direct, 1778 indirect). High blast radius increases regression risk.

---

Recommended tests

5 test(s) with exact coverage of 2 impacted unit(s). 4 impacted unit(s) have no covering tests in the selected set.

Confidence: 5 exact (5 tests selected)

Test	Confidence	Why
internal/reporting/analyze_report_test.go	exact	exact coverage of RenderAnalyzeReport
internal/reporting/analyze_report_v2_test.go	exact	exact coverage of RenderAnalyzeReportV2
internal/testdata/e2e_analysis_test.go	exact	exact coverage of RenderAnalyzeReport
internal/testdata/e2e_workflows_test.go	exact	exact coverage of RenderAnalyzeReport
internal/testdata/golden_test.go	exact	exact coverage of RenderAnalyzeReport

---

Owners: PMCLSF

Limitations

• No coverage artifacts provided; protection gaps reflect missing data, not measured absence. Provide --coverage to improve accuracy.
• Mixed test cultures reduce cross-framework optimization confidence. Consider standardizing on fewer frameworks.

_{Generated by Terrain · terrain pr --json for machine-readable output}

Targeted Test Results

Terrain selected 5 test(s) instead of the full suite.

• Go tests: passed