feat: cheapest pipeline — gpt-4o-mini-transcribe + gpt-5.4-nano + TTS

Simple 3-step chat completions pipeline at ~/usr/bin/bash.019/min total.
Streams PCM16 audio from frontend, transcribes on release,
generates response via gpt-5.4-nano, speaks via OpenAI TTS.

Cost breakdown:
  gpt-4o-mini-transcribe: /usr/bin/bash.003/min
  gpt-5.4-nano:          ~/usr/bin/bash.001/min
  OpenAI TTS (nova):     /usr/bin/bash.015/min
  Total:                 ~/usr/bin/bash.019/min (~/usr/bin/bash.57/day at 30min)

frontend/src/hooks/useConversation.ts

@@ -272,6 +272,11 @@ export function useConversation() {
     streamRef.current?.getTracks().forEach((t) => t.stop());
     streamRef.current = null;
     setIsRecording(false);
+
+    // Tell backend to process accumulated audio
+    if (wsRef.current?.readyState === WebSocket.OPEN) {
+      wsRef.current.send(JSON.stringify({ type: 'transcribe' }));
+    }
   }, []);
 
   // ── Text ──