Add Sport Cub Monte Carlo failure-envelope study

[dbansal02]

Summary

Adds a reproducible Sport Cub Monte Carlo failure-envelope study with controller, receiver, mission-condition, and physical-parameter sweeps; explicit hard-failure classification; raw CSV evidence; professional plots; and a structured HTML report.

Spec / MLS Alignment

• Checked SPEC_0021, SPEC_0025, and SPEC_0032.
• No Rumoca compiler or Modelica language-semantics change.
• Change is confined to the Sport Cub Python simulation example and evidence artifacts.

Risk and Design Notes

• Correctness risk: conclusions are valid for the public reference model, tested mission, uncertainty distributions, and parameter ranges—not direct proof for an event aircraft.
• Maintenance risk: multiple focused study entry points; they share the same model, deterministic seed, common-random trial design, and failure classification.
• No Rust public API or crate-boundary change.

Testing

• Python scripts compile successfully.
• cargo fmt --check
• cargo clippy --workspace --all-targets --all-features -- -D warnings
• cargo test --workspace
• cargo doc --no-deps
• Independently recomputed report counts and medians from raw CSVs.
• Validated all local HTML image and evidence links.
• MSL gate not run because compiler and Modelica semantic implementation are unchanged.

Code Size Budget

• production_lines_added: 3876
• production_lines_deleted: 0
• test_lines_added: 0
• test_lines_deleted: 0
• public_items_added: 0
• public_items_removed: 0
• files_touched: 35
• net_added_lines: 10399

Net growth consists of the standalone simulation model, focused experiment runners, reproducible raw evidence, plots, and report. The first compression pass reused shared trial construction, metrics, failure classification, and model code instead of duplicating those mechanisms. Further studies should extract shared campaign/plot aggregation helpers rather than add another independent framework.

Reviewer Checklist

• Relevant active specs were checked.
• Crate boundaries and phase ownership preserved.
• Tests prove behavior and standard CI gates pass.
• Size-budget section completed.
• Positive net diff has explicit compression justification.
• New APIs are required and minimal.
• No old/new compatibility paths introduced.
• No clippy allowances added.
• Commit is signed off; no AI co-author line.
• External source provenance is recorded in the example model.