feat: extract in-process MCP action layer, concurrent task dispatch, and spawn/reconcile hardening

- Extract device action execution into artemis/mcp action layer:
  typed ActionResult, capability manifest with startup validation,
  AdbActuator as the single physical implementation, in-memory
  ActionSession (replaces stdio subprocess MCP in validator), and
  McpActionExecutor preserving the legacy executor contract.
  adb_server keeps its external stdio/SSE surface unchanged and now
  reuses the actuator implementation; import-time side effects moved
  into configure_stdio_mode().
- Capability-gated prompt assembly (prompt_assembly.py): tool enums and
  teaching segments render per available backend tools; full-set output
  is byte-identical to the previous static text (locked by tests).
- Admin console task queue: serial worker replaced by a per-device
  concurrent dispatcher (ARTEMIS_CONCURRENCY_MODE=global restores
  serial semantics), per-session cancellation, strict device binding on
  enqueue, unified stdio piping with tee to trace stdout.log, and
  shutdown cleanup of all active runs. Fix _pid_exists exception order
  so PermissionError is treated as alive.
- mcp_server reliability: spawn watchdog kills runners stuck before
  interpreter startup and releases device reservations; task state
  reconcile uses psutil liveness, prefers runner-reported pids, and
  adds a 45s startup grace to stop false failure reports on Windows;
  conversation_id is now optional end-to-end; explicit device_serial is
  validated at submission.
- CLI install now writes ~/.claude/rules/artemis.md for Claude Code and
  migrates/removes the legacy CLAUDE.md injection block; ~/.claude.json
  is never rewritten when it fails to parse.
- Tests: Windows-portable stdio lifecycle and path assertions; new
  suites for action layer, prompt assembly, spawn watchdog, and task
  reconcile (895 passed, 3 skipped).

apps/admin_console/server.py

@@ -147,6 +147,23 @@ async def on_shutdown():
             pass
     state.worker_task = None
 
+    # Cancel in-flight run coroutines and wait for their finalizers (DB status,
+    # session_ended broadcast, trace sync, recording recovery) to run.
+    run_tasks = [t for t in list(task_queue_service._run_tasks) if not t.done()]
+    for run_task in run_tasks:
+        run_task.cancel()
+    if run_tasks:
+        try:
+            await asyncio.wait_for(
+                asyncio.gather(*run_tasks, return_exceptions=True), timeout=5.0
+            )
+        except (TimeoutError, asyncio.CancelledError):
+            pass
+    for run in list(state.active_runs.values()):
+        run_proc = run.get("process")
+        if run_proc is not None and run_proc.returncode is None:
+            await task_queue_service._terminate_worker_process(run_proc)
+    state.active_runs.clear()
     if state.current_process is not None and state.current_process.returncode is None:
         await task_queue_service._terminate_worker_process(state.current_process)
     for item in state.queue_items: