Improve mutation channel robustness

- Add event type constants (MutationCreate, MutationUpdate, MutationDelete, MutationComment)
- Make buffer size configurable via BEADS_MUTATION_BUFFER (default 512, up from 100)
- Add defensive closed-channel handling in event loop consumer
- Add 1s dropped-events ticker (down from 60s) for faster recovery
- Verify mutationChan is never closed (prevents panic)

Oracle review findings addressed:
- Eliminates panic risk from send-on-closed-channel
- Reduces worst-case recovery latency from 60s to 1s
- Increases buffer capacity for better burst handling
- Type-safe event constants prevent string typos

Related: bd-36320a04, bd-1f4086c5

cmd/bd/daemon_event_loop.go

@@ -60,7 +60,12 @@ func runEventDrivenLoop(
 	go func() {
 		for {
 			select {
-			case event := <-mutationChan:
+			case event, ok := <-mutationChan:
+				if !ok {
+					// Channel closed (should never happen, but handle defensively)
+					log.log("Mutation channel closed; exiting listener")
+					return
+				}
 				log.log("Mutation detected: %s %s", event.Type, event.IssueID)
 				exportDebouncer.Trigger()
 
@@ -70,23 +75,28 @@ func runEventDrivenLoop(
 		}
 	}()
 
-	// Optional: Periodic health check and dropped events safety net
+	// Periodic health check
 	healthTicker := time.NewTicker(60 * time.Second)
 	defer healthTicker.Stop()
 
+	// Dropped events safety net (faster recovery than health check)
+	droppedEventsTicker := time.NewTicker(1 * time.Second)
+	defer droppedEventsTicker.Stop()
+
 	for {
 		select {
-		case <-healthTicker.C:
-			// Periodic health validation (not sync)
-			checkDaemonHealth(ctx, store, log)
-			
-			// Safety net: check for dropped mutation events
+		case <-droppedEventsTicker.C:
+			// Check for dropped mutation events every second
 			dropped := server.ResetDroppedEventsCount()
 			if dropped > 0 {
 				log.log("WARNING: %d mutation events were dropped, triggering export", dropped)
 				exportDebouncer.Trigger()
 			}
 
+		case <-healthTicker.C:
+			// Periodic health validation (not sync)
+			checkDaemonHealth(ctx, store, log)
+
 		case sig := <-sigChan:
 			if isReloadSignal(sig) {
 				log.log("Received reload signal, ignoring")