feat: Sidecar V2 — multi-pool provider proxy with 429 cooldown

- proxy.py: Fix route path duplication (v1/v1 → v1) when upstream
  base URL already includes /v1 prefix
- proxy.py: Fix _emergency_count global variable for metrics tracking
- server.py: Add logging.basicConfig(level=logging.INFO) for structlog
  INFO-level log visibility
- Full multi-pool routing: primary → fallback → emergency passthrough
- Per-backend rate limiting with RPM-based token bucket
- 429 cooldown mechanism with automatic recovery
- Dashboard with SSE real-time monitoring
- Admin API for backend/pool/config management
- SQLite-backed persistence with encrypted API key storage
- Docker compose deployment

Deployed by opengineer 严维序 as BIZ-50 Step 4

router.py

@@ -0,0 +1,62 @@
+"""Model → Backend routing logic for Sidecar V2."""
+
+import structlog
+from typing import Optional
+
+from storage.models import Backend
+from pool_manager import PoolManager
+from rate_limiter import PerBackendRateLimiter
+
+logger = structlog.get_logger("sidecar_v2.router")
+
+
+class Router:
+    """Routes model requests to the best available backend.
+
+    Pick strategy:
+    1. Primary pool → healthy backends supporting the model
+    2. Rate-limiter check → skip if RPM exhausted
+    3. Fallback pool → repeat above
+    4. If all exhausted → return None (caller handles emergency)
+    """
+
+    def __init__(self, pool_manager: PoolManager, rate_limiter: PerBackendRateLimiter):
+        self._pool_manager = pool_manager
+        self._rate_limiter = rate_limiter
+
+    def pick_backend(self, canonical_model: str) -> Optional[Backend]:
+        """Pick the best available backend for a model.
+
+        Tries primary pool first, then fallback.
+        Within each pool, skips backends at RPM limit.
+        Returns None if no backend available.
+        """
+        # Try pools in order
+        for pool in ["primary", "fallback"]:
+            backends = self._pool_manager.get_available_backends(
+                canonical_model, pool=pool
+            )
+            for backend in backends:
+                # Rate-limit check
+                if self._rate_limiter.consume(
+                    backend.id, backend.rpm_limit
+                ):
+                    return backend
+                # Skip this backend, try next
+                logger.debug(
+                    "backend_rate_limited",
+                    backend_id=backend.id,
+                    pool=pool,
+                    model=canonical_model,
+                )
+
+            if not backends:
+                logger.debug("pool_exhausted", pool=pool, model=canonical_model)
+            else:
+                logger.debug("pool_rpm_exhausted", pool=pool, model=canonical_model)
+
+        return None
+
+    def get_all_pools_exhausted_info(self, canonical_model: str) -> bool:
+        """Check if ALL pools are exhausted for a model."""
+        return not self._pool_manager.is_any_pool_available(canonical_model)