Feat v0.6.0 cluster registry + HA

PG-backed cluster registry for horizontal scaling — zero new
infrastructure (no etcd/Consul/Redis). MVP convergence point.

- node_registry UNLOGGED table (migration 013)
- Self-registration, heartbeat tick (10s), stale sweep (30s),
  self-eviction (os.Exit on 0 rows → K8s restarts)
- GET /api/v1/admin/cluster returns nodes with runtime stats
- Health endpoints include node_id + cluster size/peers
- cluster-dashboard admin surface with auto-refresh
- 3-replica E2E test (docker-compose-e2e.yml + ci/e2e-cluster-test.sh)
- chat + cluster-dashboard added to curated default bundle
- 5 new tests (3 unit + 2 handler), all passing

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

packages/cluster-dashboard/manifest.json

@@ -0,0 +1,13 @@
+{
+  "id": "cluster-dashboard",
+  "icon": "🔗",
+  "type": "surface",
+  "title": "Cluster",
+  "route": "/s/cluster-dashboard",
+  "auth": "admin",
+  "layout": "single",
+  "components": [],
+  "hooks": ["surface"],
+  "version": "0.6.0",
+  "description": "Cluster node registry — health, runtime stats, and peer visibility."
+}