Add weighted readiness scoring engine and team readiness API

[Copilot]

This PR introduces the core readiness scoring flow requested in the issue: compute athlete readiness from recent audit logs (0–100) and expose team-level readiness as the average of member scores.

• Core scoring engine

  • Added backend/readiness_engine.py to compute athlete readiness from the latest 3 daily logs.
  • Implemented weighted categories per requirement:
    • Physical (rpe, fatigue) → 40%
    • Mental (stress, sleep) → 40%
    • Logistics + Environment (logistics, environment) → 20%
  • Added team aggregation utilities to average per-athlete readiness into a team score.

• Team readiness API

  • Added endpoint: GET /api/v1/teams/{team_id}/readiness
  • Endpoint returns:
    • team metadata (team_id, team_name, member_count)
    • readiness_score (team average)
    • per-athlete readiness breakdown
  • Returns 404 when the team is not found.

• Readiness data source

  • Added backend/data/readiness_audit_logs.json containing team membership and audit logs used by the scoring engine.

• Focused backend coverage

  • Added backend/tests/test_readiness_engine.py with tests for:
    • latest-3-days selection behavior
    • weighted averaging across varying daily logs
    • team-score aggregation from member scores

# readiness_engine.py
score = (
    (physical * 0.4) +
    (mental * 0.4) +
    (contextual * 0.2)
)

@app.get("/api/v1/teams/{team_id}/readiness")
async def team_readiness(team_id: str):
    readiness = get_team_readiness(team_id)
    if not readiness:
        raise HTTPException(status_code=404, detail=f"Team {team_id} not found")
    return readiness