feat: async sandbox lifecycle with Redis Stream events

Replace synchronous RPC-based CP-host communication for sandbox
lifecycle operations (Create, Pause, Resume, Destroy) with an async
pattern. CP handlers now return 202 Accepted immediately, fire agent
RPCs in background goroutines, and publish state events to a Redis
Stream. A background consumer processes events as a fallback writer.

Agent-side auto-pause events are pushed to the CP via HTTP callback
(POST /v1/hosts/sandbox-events), keeping Redis internal to the CP.

All DB status transitions use conditional updates
(UpdateSandboxStatusIf, UpdateSandboxRunningIf) to prevent race
conditions between concurrent operations and background goroutines.

The HostMonitor reconciler is kept at 60s as a safety net, extended
to handle transient statuses (starting, pausing, resuming, stopping).

Frontend updated to handle 202 responses with empty bodies and render
transient statuses with blue indicators.

pkg/cpserver/run.go

@@ -187,8 +187,13 @@ func Run(opts ...Option) {
 	// Start channel event dispatcher.
 	channelDispatcher.Start(ctx)
 
-	// Start host monitor (passive + active reconciliation every 30s).
-	monitor := api.NewHostMonitor(queries, hostPool, al, 15*time.Second)
+	// Start sandbox event consumer (processes lifecycle events from Redis stream).
+	sandboxEventConsumer := api.NewSandboxEventConsumer(rdb, queries, al)
+	sandboxEventConsumer.Start(ctx)
+
+	// Start host monitor (passive + active reconciliation every 60s).
+	// Reduced from 15s since async events handle the normal case.
+	monitor := api.NewHostMonitor(queries, hostPool, al, 60*time.Second)
 	monitor.Start(ctx)
 
 	// Hard-delete accounts that have been soft-deleted for more than 15 days (runs every 24h).