feat(dashboard): per-stage STT/LLM/TTS timing + rename 두뇌 -> LLM

- voice_turn now records three separate timed steps (STT, LLM, TTS) instead of
  one combined "STT+두뇌+TTS" step, so each stage's latency shows in the turn.
- Rename 두뇌 -> LLM everywhere user-facing: component chip, sub header, demo
  banner, model label, brain-failure log, startup print, pipeline step name.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

wsai/__main__.py

@@ -142,7 +142,7 @@ def _run_voice_server(host: str, port: int) -> None:
     monitor.log("info", f"음성 서버 준비 완료 (STT device={sdev}). 디스코드 봇 연결 대기.", cat="READY")
 
     shown = host if host not in ("0.0.0.0", "") else _lan_ip()
-    print(f"\n  음성 서버 준비 완료 (STT device: {sdev}, 두뇌: {brain_name})")
+    print(f"\n  음성 서버 준비 완료 (STT device: {sdev}, LLM: {brain_name})")
     print(f"  대시보드/상태:   http://{shown}:{port}")
     print(f"  봇 연결 엔드포인트: http://127.0.0.1:{port}/api/voice-turn\n")
     try: