feat: NanoClaude Phase 2 — EXAONE→Gemma 파이프라인, 큐, 상태 API

- ModelAdapter: 범용 OpenAI-compat 어댑터 (stream/complete/health)
- BackendRegistry: rewriter(EXAONE) + reasoner(Gemma4) 헬스체크 루프
- 2단계 파이프라인: EXAONE rewrite → Gemma reasoning (SSE rewrite 이벤트 노출)
- Fallback: 맥미니 다운 시 EXAONE 단독 모드, stream 중간 실패 시 자동 전환
- Cancel-safe: rewrite 전/후, streaming loop 내, fallback 경로 모두 체크
- Rewrite heartbeat: complete_chat 대기 중 2초 간격 processing 이벤트
- JobQueue: Semaphore(3) 기반 동시성 제한, 정확한 queue position
- GET /chat/{job_id}/status, GET /queue/stats 엔드포인트
- DB: rewrite_model, reasoning_model, rewritten_message 컬럼 추가

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

nanoclaude/main.py

@@ -1,4 +1,4 @@
-"""NanoClaude — 비동기 job 기반 AI Gateway."""
+"""NanoClaude — 비동기 job 기반 AI Gateway (Phase 2: EXAONE → Gemma 파이프라인)."""
 
 from __future__ import annotations
 
@@ -12,22 +12,29 @@ from fastapi.responses import JSONResponse
 from config import settings
 from db.database import init_db
 from routers import chat
+from services.backend_registry import backend_registry
+from services.job_queue import init_queue, job_queue
 
 logging.basicConfig(
     level=logging.INFO,
     format="%(asctime)s %(levelname)s %(name)s — %(message)s",
 )
 
+
 @asynccontextmanager
 async def lifespan(app: FastAPI):
     await init_db()
+    backend_registry.init_from_settings(settings)
+    backend_registry.start_health_loop(settings.health_check_interval)
+    init_queue(settings.max_concurrent_jobs)
     yield
+    backend_registry.stop_health_loop()
 
 
 app = FastAPI(
     title="NanoClaude",
-    version="0.1.0",
-    description="비동기 job 기반 AI Gateway — Phase 1",
+    version="0.2.0",
+    description="비동기 job 기반 AI Gateway — Phase 2 (EXAONE → Gemma 파이프라인)",
     lifespan=lifespan,
 )
 
@@ -54,9 +61,14 @@ app.include_router(chat.router)
 
 @app.get("/")
 async def root():
-    return {"service": "NanoClaude", "version": "0.1.0", "phase": 1}
+    return {"service": "NanoClaude", "version": "0.2.0", "phase": 2}
 
 
 @app.get("/health")
 async def health():
-    return {"status": "ok"}
+    from services.job_queue import job_queue
+    return {
+        "status": "ok",
+        "backends": backend_registry.health_summary(),
+        "queue": job_queue.stats if job_queue else {},
+    }