Persist the PM-respawn counter across orchestrator restarts (#275)

* feat(orchestrator): add respawn_tracker table + migration 051

Durable backing for AgentOrchestrator._pm_respawn_tracker (the PM-respawn
loop breaker). Kept only in memory it reset to count=1 on every restart,
re-burning the strike threshold against a still-wedged task. RespawnTrackerTable
mirrors WaitingRecordTable: composite PK (agent_slug, task_id) matching the
in-memory key; task_id is intentionally NOT a FK (the startup loader validates
against live tasks so a stale counter can't resurrect). Migration 051 verified
with a real alembic upgrade head + downgrade -1 + re-upgrade on Postgres.

* feat(orchestrator): persist the PM-respawn counter across restarts

The PM-respawn loop breaker (_pm_respawn_tracker) lived only in memory, so an
orchestrator restart reset a wedged task's strike count to 1 and re-burned the
whole threshold (4 spawns x container cost) before the gate fired again.

Write-through each gate mutation to the respawn_tracker table via a
fire-and-forget _schedule_respawn_persist (on the existing _bg_tasks strong-ref
set; a DB hiccup degrades to in-memory-only, never gates/un-gates a spawn), and
restore_respawn_tracker() repopulates the counter at startup, validating each
row against live tasks (drops terminal/missing) so a stale counter can't
resurrect against a fixed task. Best-effort + inert when the table is empty.
Cannot manufacture a spawn — the counter only ever suppresses one.

(_instances reconcile, the spec's other goal, already shipped as
_readopt_running_agents.)

* fix(types): cast Mapped[UUID] columns in project routes + self_heal

A clean `mypy roboco/ tests/` run surfaces 7 pre-existing errors in files this
branch doesn't touch: project-route handlers and self_heal_engine pass a
ProjectTable.id (declared Mapped[UUID] against SQLAlchemy's dialect UUID, so
mypy infers sqlalchemy.sql.sqltypes.UUID[Any]) where a uuid.UUID is expected.
An incremental .mypy_cache had hidden them. Apply the same targeted cast unblock
used for the prior batch; the deeper fix (migrating the ~88 Mapped[UUID]
columns to Mapped[uuid.UUID]) remains a separate dedicated task.

* docs(orchestrator): document respawn_tracker durability

Add the orchestrator runtime-state durability note to CLAUDE.md (respawn_tracker
write-through + restore; _instances reconciled-from-Docker) + the migration-051
narrative, and a CHANGELOG [Unreleased] Fixed entry. Also type-clean the new
respawn_tracker table test (cast __table__ to Table under TYPE_CHECKING).

---------

Co-authored-by: Renn F <rennf93@users.noreply.github.com>

CLAUDE.md

@@ -512,6 +512,7 @@ Server-side events reach these sockets through `roboco/api/websocket_bridge.py`,
 - **Provider overloads** reuse the same park-and-probe break. A persistent model-API overload (HTTP 529 / 500 / 503 — the SDK already retries transient ones) parks the provider exactly like a 429 instead of crash-retrying the agent straight back into the overload and burning tokens; the overload is detected orchestrator-side from the dead container's log markers, and the background loop revives the parked work when it recovers. The same break also catches the **Claude session-limit** 429 (the org's 5-hour usage window): an agent exiting with a 0-token session-limit rejection parks the provider and is auto-revived when the window resets, instead of fleet-wide crash-respawning straight back into the limit. Gated by `ROBOCO_OVERLOAD_BREAK_ENABLED` (default-on).
 - **Gateway-health recovery** closes a blind spot in the stale-claim reaper: the heartbeat is bumped only by gateway verbs, so a broken-but-alive agent (a corrupted `/app/.venv` so no gateway tool imports) goes heartbeat-stale yet keeps its container up, and the reaper's live-skip would protect it forever. On a stale-heartbeat live container the reaper now probes the gateway out-of-band (`_probe_gateway_health` → `docker exec` the gateway venv imports) and, once broken past `ROBOCO_GATEWAY_HEALTH_GRACE_SECONDS` (a transient probe miss is tolerated), kills + evicts it (`_maybe_recover_broken_gateway`) so it falls through to release + respawn; healthy or inconclusive probes spare it. Gated by `ROBOCO_GATEWAY_HEALTH_ENABLED` (default-on). It is the third leg beside the shipped bash-guard `/app` block (prevents the self-corruption) and the reaper Docker-liveness fallback (stops over-reaping live containers).
 - **PM coordinator concurrency.** A Main / Cell PM plans and delegates many root tasks in parallel — the actual work then runs in the delegated children/cells, not in the PM's own hands. The claim-time concurrency guards that keep a *developer* to one task at a time (`already_active` / `paused`, in `roboco/services/gateway/claim_guards.py`) are therefore **skipped for the coordinator PM roles** (`_COORDINATOR_ROLES = {main_pm, cell_pm}`, consulted in `_run_claim_guards`); only a genuine upstream **sequence dependency** (`unmet_dependency`, which parks the task back to `pending`) holds a PM's root back. Without this a single PM that claimed one root could never plan a second — it thrashed between its claimed roots and respawned forever, burning tokens for zero progress (the live `i_am_idle`-auto-paused-umbrella deadlock). The `paused` guard also excludes the target task itself, so a PM re-entering its own paused umbrella never self-blocks.
+- **Orchestrator runtime-state durability.** The PM-respawn loop breaker (`_pm_respawn_tracker`, the `(agent_slug, task_id) → strike-count` circuit breaker) is **DB-durable** via the `respawn_tracker` table (migration 051): each gate mutation write-throughs fire-and-forget on the `_bg_tasks` set (`_schedule_respawn_persist` → `_persist_respawn_record`), and `restore_respawn_tracker()` repopulates it at `start()`, validating each row against live tasks (terminal/missing rows are evicted). Kept only in memory it reset to `count=1` on every restart and re-burned the whole strike threshold (4 spawns) against a still-wedged task. It mirrors the `WaitingRecordTable` / `restore_waiting_records` pattern: best-effort (a DB hiccup degrades to in-memory-only — it can only ever *suppress* a spawn, never manufacture one) and inert when the table is empty. The companion `_instances` registry is **reconciled-from-Docker** (not persisted) at startup via `_readopt_running_agents`, so the reaper's liveness path and the spawn gate's `_is_agent_active` check see surviving containers immediately after a restart.
 - **Token usage** is captured per agent session from the Claude Code transcript via the SDK server's `/usage/sync` (hook → orchestrator finalize → `agent_spawn_sessions` → `daily_usage_rollups` → dashboard). Cost uses provider-aware pricing in `roboco/billing/pricing.py` (Anthropic priced; local/Ollama intentionally `$0`). The token sweep also publishes `USAGE_SNAPSHOT` to `/ws/system`, so the dashboard's "Token Usage & Cost" panel updates live and falls back to HTTP polling when the stream is down.
 - **Delivery observability** (the panel's Metrics → "Delivery" tab) shows how work *flows*, computed by `MetricsService` from data already captured — no new feature flag. Per-stage cycle time and the bottleneck distribution are reconstructed from the `audit_log` transition journey (each generic `task.<status>` event marks entry into a status; the named `task.qa_fail`/`task.pr_fail` events are excluded from the reconstruction). Rework rate reads `tasks.revision_count` — incremented once per transition into `needs_revision` at the single chokepoint `TaskService._emit_status_transition_audit` — and attributes each bounce to the QA / PR-reviewer via those named audit events; rework cost joins `agent_spawn_sessions.task_id`. Read-only endpoints: `/dashboard/metrics/{cycle-time,bottlenecks,rework,scorecard/agent/{id},scorecard/team/{team}}`.