Address PR review feedback: fix markdown tables, typo, and vehicle cleanup

[Copilot]

Addresses all review comments from PR #6, fixing markdown rendering issues, a typo, and adding proper resource cleanup to the probe script.

Markdown table fixes (docs/metrics_catalog.md)

• Removed extra empty columns from Vehicle State and Collision Event tables
• Fixed broken inline code blocks split across columns (e.g., vehicle.speed and event.intensity)
• Standardized "computed magnitude" annotation format

Typo correction (src/cot/carla_client.py)

• Fixed "dependant" → "dependent" in comment

Vehicle cleanup (src/scripts/probe_world_and_vehicle.py)

• Added try/finally block to destroy spawned vehicles, preventing actor accumulation across runs
• Added synchronous mode detection to use world.tick() vs world.wait_for_tick() appropriately

try:
    v, spawned = get_or_spawn_vehicle(world)
    # ... probe metrics ...
finally:
    if spawned and v is not None and getattr(v, "is_alive", False):
        v.destroy()

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.