feat: Hub monitoring takeover — event system, dead man's switch, notifications (v0.3.0)

Replace external Healthchecks.io with Hub-native monitoring. New events
table + /api/v1/event endpoint for structured events from controllers.
Staleness checker (60s) detects unresponsive nodes. Backup deadline
checker (daily 05:00) catches missed backups. Notification dispatcher
sends operator (English) + customer (Hungarian) emails via Resend with
per-event cooldowns. Event timeline on customer page, dashboard badges.
Config form deprecates Monitoring UUIDs section.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

hub/internal/web/controller.yaml.default

@@ -80,12 +80,12 @@ backup:
 monitoring:
   enabled: true
   healthchecks_base: "https://status.felhom.eu"
-  ping_uuids:
-    heartbeat: ""                                  # Every 5 min — controller process alive
-    system_health: ""                              # Every 5 min — comprehensive system check
-    db_dump: ""                                    # Daily — after database dumps
-    backup: ""                                     # Daily — after restic snapshot
-    backup_integrity: ""                           # Weekly (Sunday) — restic check
+  # ping_uuids: (deprecated — monitoring is now handled by the Hub event system)
+  #   heartbeat: ""
+  #   system_health: ""
+  #   db_dump: ""
+  #   backup: ""
+  #   backup_integrity: ""
   system_health_interval: "5m"
   health_check_schedule: "06:00"
   thresholds: