fix(scheduler): mark orphaned live assets error immediately when recorder stops

When a capture sidecar crashes before finalize() runs (e.g. wrong node,
filter error, hardware fault), the asset stays 'live' indefinitely — library
shows 'Recording' badge for up to 120 minutes until the stale-timeout fires.

Add an orphan check that runs every scheduler tick: if an asset is 'live'
and its recorder is 'stopped', mark it 'error' immediately. This runs before
the 120-minute staleness guard so the library clears within 15 seconds.

🤖 Generated with Claude Code

services/mam-api/src/scheduler.js

@@ -135,6 +135,24 @@ async function tick() {
       }
     }
 
+    // Orphaned live assets: recorder stopped but asset still 'live'.
+    // Happens when the capture sidecar crashes before finalize() runs.
+    // Mark error immediately so the library doesn't show "Recording" forever.
+    const orphanResult = await client.query(
+      `UPDATE assets a
+       SET status = 'error', updated_at = NOW()
+       FROM recorders r
+       WHERE a.status = 'live'
+         AND a.display_name = r.current_session_id
+         AND r.status = 'stopped'
+       RETURNING a.id, a.display_name`
+    );
+    if (orphanResult.rows.length > 0) {
+      for (const row of orphanResult.rows) {
+        console.warn(`[scheduler] orphaned live asset (recorder stopped): ${row.id} (${row.display_name})`);
+      }
+    }
+
     const LIVE_TIMEOUT_MINUTES = parseInt(process.env.LIVE_ASSET_TIMEOUT_MINUTES || '120', 10);
     const staleResult = await client.query(
       `UPDATE assets