fix(heartbeat): prevent false process_lost failures on queued and non-child-process runs

- reapOrphanedRuns() now only scans running runs; queued runs are
  legitimately absent from runningProcesses (waiting on concurrency
  limits or issue locks) so including them caused false process_lost
  failures (closes #90)
- Add module-level activeRunExecutions set so non-child-process adapters
  (http, openclaw) are protected from the reaper during execution
- Add resumeQueuedRuns() to restart persisted queued runs after a server
  restart, called at startup and each periodic tick
- Add outer catch in executeRun() so setup failures (ensureRuntimeState,
  resolveWorkspaceForRun, etc.) are recorded as failed runs instead of
  leaving them stuck in running state
- Guard resumeQueuedRuns() against paused/terminated/pending_approval agents
- Increase opencode models discovery timeout from 20s to 45s

server/package.json

@@ -38,9 +38,9 @@
     "@paperclipai/adapter-codex-local": "workspace:*",
     "@paperclipai/adapter-cursor-local": "workspace:*",
     "@paperclipai/adapter-gemini-local": "workspace:*",
+    "@paperclipai/adapter-openclaw-gateway": "workspace:*",
     "@paperclipai/adapter-opencode-local": "workspace:*",
     "@paperclipai/adapter-pi-local": "workspace:*",
-    "@paperclipai/adapter-openclaw-gateway": "workspace:*",
     "@paperclipai/adapter-utils": "workspace:*",
     "@paperclipai/db": "workspace:*",
     "@paperclipai/shared": "workspace:*",