feat(mqtt): add device health telemetry and migrate to v2 topic spec

Two efforts that landed together because the v2 topic work extends
tables the health-telemetry effort added days earlier in the same
files/functions, making them impractical to separate cleanly:

Health/diagnostics telemetry (schema, Jul 13-17):
- New Postgres tables: device_alert_events, device_boot_events,
  device_ping_samples, device_diagnostics_reports, plus a `source`
  column on device_logs to distinguish log origins
- Query/service layer in pg_mqtt.py and database/__init__.py for
  inserting and listing this history, plus a "latest metrics" endpoint
  combining most-recent diagnostics + ping RTT per device
- mqtt/router.py gains list endpoints for alert/boot/ping/diagnostics
  history, consumed by the upcoming Health tab

MQTT v2 topic migration (Sep 21):
- Heartbeat payload flattened per vesper_mqtt_topic_spec_v2.md, adding
  rssi/free_heap/state/ok fields
- Command replies move to control/ack, device-initiated events to
  control/reports; mqtt/client.py subscribes to the new topic set and
  runs a ping_loop (wired up in main.py) for RTT sampling
- mqtt/logger.py and pg_mqtt.py updated to parse and persist the new
  payload shape alongside the legacy fields for backwards compatibility

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

backend/mqtt/client.py

@@ -1,12 +1,19 @@
 import json
 import logging
 import asyncio
+import time
 from typing import Set
 import paho.mqtt.client as paho_mqtt
 from config import settings
 
 logger = logging.getLogger("mqtt.client")
 
+PING_INTERVAL_SECONDS = 60
+# Only devices heard from within this window get pinged — no point spending
+# broker traffic/RTT samples on a device that's already known offline; its
+# heartbeat-derived "online" state will already reflect that on the console.
+PING_ONLINE_WINDOW_SECONDS = 90
+
 
 class MqttManager:
     """Singleton MQTT client manager."""
@@ -61,12 +68,17 @@ class MqttManager:
         if reason_code == 0:
             self._connected = True
             logger.info("MQTT connected, subscribing to topics")
+            # v2 topic set — see vesper_mqtt_topic_spec_v2.md in the firmware repo.
+            # control/command is inbound-to-device only; the console never subscribes to it.
             client.subscribe([
-                ("vesper/+/data", 1),
+                ("vesper/+/control/ack", 1),
+                ("vesper/+/control/reports", 1),
                 ("vesper/+/status/heartbeat", 1),
-                ("vesper/+/status/alerts", 1),
-                ("vesper/+/status/info", 0),
-                ("vesper/+/logs", 1),
+                ("vesper/+/status/playback", 1),
+                ("vesper/+/system/alerts", 1),
+                ("vesper/+/system/info", 1),
+                ("vesper/+/system/logs", 0),
+                ("vesper/+/system/metrics", 0),
             ])
         else:
             logger.error(f"MQTT connection failed: {reason_code}")
@@ -132,10 +144,48 @@ class MqttManager:
         if not self._client or not self._connected:
             return False
 
-        topic = f"vesper/{device_serial}/control"
-        payload = json.dumps({"cmd": cmd, "contents": contents})
+        topic = f"vesper/{device_serial}/control/command"
+        payload = json.dumps({"v": 2, "cmd": cmd, "contents": contents})
         result = self._client.publish(topic, payload, qos=1)
         return result.rc == paho_mqtt.MQTT_ERR_SUCCESS
 
+    async def ping_loop(self):
+        """Periodically pings every recently-online device with a client
+        timestamp so mqtt/logger.py::_handle_data_response can compute RTT
+        from the echoed pong. Deliberately bypasses db.insert_command — this
+        is a background health check, not a user-initiated command, and
+        shouldn't clutter the Control tab's command history.
+
+        Only available on RTC-equipped firmware builds (see API Reference —
+        ping is compiled out on agnus/agnus-mini). A pong simply never
+        arrives for those devices, so no ping-latency samples accumulate for
+        them; the Health tab handles an empty series as "unsupported".
+        """
+        while True:
+            await asyncio.sleep(PING_INTERVAL_SECONDS)
+            try:
+                await self._ping_online_devices()
+            except Exception as e:
+                logger.error(f"Ping loop error: {e}")
+
+    async def _ping_online_devices(self):
+        import database as db
+        heartbeats = await db.get_latest_heartbeats()
+        now = time.time()
+        for hb in heartbeats:
+            try:
+                from datetime import datetime
+                received = datetime.fromisoformat(hb["received_at"])
+                age = now - received.timestamp()
+            except (ValueError, TypeError, KeyError):
+                continue
+            if age > PING_ONLINE_WINDOW_SECONDS:
+                continue
+            self.publish_command(
+                device_serial=hb["device_serial"],
+                cmd="ping",
+                contents={"ts": int(now * 1000)},
+            )
+
 
 mqtt_manager = MqttManager()