Implement pluggable fire-spotting models from CA comparison study

[Copilot]

Adds four spotting distance models (Alexandridis, Trucchia, Pereira, Koo) from the 2024 Agricultural and Forest Meteorology comparison study, following the existing ROS model selection architecture.

Implementation

• Model functions: Four JIT-compiled spotting distance calculators in functions.py, each with different physics (exponential, power-law, log-normal, flight-time based)
• Factory pattern: get_spotting_fn() selector matching get_p_time_fn() design
• Propagation chain: Threaded spotting_fn parameter through compute_spotting() → single_cell_updates() → next_updates_fn() → advance_front_until()
• Configuration: Added spotting_model field to PropagatorConfigurationLegacy with default "alexandridis"
• Type system: SpottingModel literal type for compile-time validation

Usage

Configuration-based:

{
  "do_spotting": true,
  "spotting_model": "trucchia"
}

Programmatic:

from propagator.core.numba import get_spotting_fn

sim = Propagator(
    dem=dem,
    veg=veg,
    do_spotting=True,
    spotting_fn=get_spotting_fn("koo")
)

Testing

Added 13 tests covering factory behavior, model physics (zero wind, directional effects, stochastic behavior), and edge cases.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• astral.sh
  • Triggering command: /usr/bin/curl curl -LsSf REDACTED (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Spotting models</issue_title>
<issue_description>Implement and test spotting models from "Fire-spotting modelling in operational wildfire simulators based on Cellular Automata: A comparison study" - link</issue_description>

<agent_instructions>Mimic how selection for different rate of spread formulation is implemented</agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes Spotting models #29

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.