What does this PR do?

Fixes the Docker/s6 gateway auto-start regression from #42675.

When the gateway receives an unexpected external SIGTERM, start_gateway() already exits non-zero so a service manager can revive it. However, the gateway teardown still persisted gateway_state="stopped". In Docker s6 deployments, container_boot only auto-starts profile gateway services whose previous state is running, so a container restart or image upgrade could leave messaging channels dark even though the gateway had been running before the supervisor stopped it.

This PR preserves gateway_state="running" for unexpected signal-initiated shutdowns while keeping intentional stops and planned restarts as stopped.

Related Issue

Fixes #42675

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✨ New feature (non-breaking change that adds functionality)
• 🔒 Security fix
• 📝 Documentation update
• ✅ Tests (adding or improving test coverage)
• ♻️ Refactor (no behavior change)
• 🎯 New skill (bundled or hub)

Changes Made

• gateway/run.py
  • Track unexpected signal-initiated shutdowns separately from planned stops/restarts.
  • Preserve the persisted runtime gateway state as running for those external signal shutdowns so Docker/s6 reconciliation can bring the gateway back up on the next container boot.
  • Keep planned stop/restart behavior unchanged.
• tests/gateway/test_gateway_shutdown.py
  • Add regression coverage that signal-preserved shutdowns persist running.
• tests/gateway/restart_test_helpers.py
  • Initialize the new runner test flag in the object-constructed test helper.

How to Test

1. Reproduce the bug condition: a running Docker/s6 gateway receives an external SIGTERM during container restart/upgrade; prior behavior writes gateway_state="stopped", so container_boot registers the service down on the next boot.
2. Verify the fix: unexpected signal shutdown now sets the preserve flag and final runtime status remains running; planned stops/restarts still use the existing paths.
3. Local checks run:
   • .venv/bin/python -m pytest -o 'addopts=' tests/gateway/test_gateway_shutdown.py -q → 15 passed in 35.06s
   • .venv/bin/python -m compileall -q gateway/run.py tests/gateway/test_gateway_shutdown.py tests/gateway/restart_test_helpers.py
   • .venv/bin/python -m ruff check gateway/run.py tests/gateway/test_gateway_shutdown.py tests/gateway/restart_test_helpers.py → All checks passed!
   • git diff --check

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: Linux container/PRoot environment with focused gateway shutdown tests

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

For New Skills

N/A

Screenshots / Logs

$ .venv/bin/python -m pytest -o 'addopts=' tests/gateway/test_gateway_shutdown.py -q
...............                                                          [100%]
15 passed in 35.06s

$ .venv/bin/python -m ruff check gateway/run.py tests/gateway/test_gateway_shutdown.py tests/gateway/restart_test_helpers.py
All checks passed!