checkin: report an unreachable secondary-sdr as null, never keep stale 'Running'

QA blocker: when the secondary-sdr container was down, the checkin
omitted secondary_sdr_running / sdr_devices / op25_sdr_serial, and C2
only overwrites keys that are present, so the dashboard kept the last
'Running' forever. Send explicit nulls. The local card also says 'not
reported' rather than 'not plugged in' for a pin it can't check.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

drb-edge-node/app/internal/mqtt_manager.py

@@ -236,6 +236,12 @@ class MQTTManager:
         }
         payload["sdr_pins"] = config.sdr_pins
         secondary = await secondary_sdr_client.status()
+        # Unreachable secondary-sdr: say so with explicit nulls. C2 only
+        # overwrites keys present in the checkin, so omitting them would leave
+        # a dead decoder showing "Running" forever.
+        payload["secondary_sdr_running"] = None
+        payload["sdr_devices"] = None
+        payload["op25_sdr_serial"] = None
         if secondary is not None:
             payload["secondary_sdr_running"] = [r["mode"] for r in secondary.get("running", [])]
             if secondary.get("devices") is not None: