feat(ai-core): trace instrumentation, LLM thinker, decay scheduler

- orchestrator: Tracer interface with spans for intent analysis, persona,
  subsession dispatch, context build, LLM streaming and each tool call;
  traces finish when the stream completes — the admin Trace waterfall
  finally receives data
- background: Thinker generates real LLM thoughts (post-chat + silence
  triggers, recent history as context, empty/overlong outputs discarded);
  thoughts broadcast to web clients as 'thinking' WS messages; pipeline
  postChatHook triggers post-chat thinking after each round
- memory: ApplyDecayAll + ListUserIDs with a 6h decay/purge cron in main
  (was implemented but never scheduled)

frontend/web/src/hooks/useWebSocket.ts

@@ -44,6 +44,14 @@ export function useWebSocket() {
           case 'device_update':
             if (msg.devices) setIotDevices(msg.devices)
             break
+          case 'thinking':
+            if (msg.content) {
+              addMessage({
+                id: crypto.randomUUID(), session_id: currentSessionId || '',
+                role: 'system', content: `💭 ${msg.content}`, created_at: new Date().toISOString(),
+              })
+            }
+            break
           case 'error':
             addMessage({ id: crypto.randomUUID(), session_id: '', role: 'system',
               content: `Error: ${msg.error}`, created_at: new Date().toISOString() })