feat: dashboard UX optimization + real-time backend stats + health probe fix + pool shuffle

- dashboard.html: major UX overhaul (+657/-308 lines)
- server.py: /api/admin/backends now returns real-time RPM and model_count
- pool_manager.py: random.shuffle backends for load distribution
- config.py: health probe endpoint /v1/models → /models
- docker-compose.yml: add SIDECAR_PRIMARY_WAIT_MAX_RETRIES=6

BIZ-52 post-review optimizations

config.py

@@ -39,7 +39,7 @@ class Config:
     # Health check
     health_check_interval_seconds: int = 60
     health_check_timeout_seconds: int = 10
-    health_probe_endpoint: str = "/v1/models"
+    health_probe_endpoint: str = "/models"
 
     # Admin auth
     admin_token: str = ""

docker-compose.yml

@@ -15,6 +15,7 @@ services:
       - SIDECAR_METRICS_PORT=9191
       - SIDECAR_DB_PATH=/app/data/sidecar_v2.db
       - SIDECAR_BACKUP_DIR=/app/data/backups
+      - SIDECAR_PRIMARY_WAIT_MAX_RETRIES=6
     volumes:
       - sidecar-data:/app/data
 

pool_manager.py

@@ -1,5 +1,6 @@
 """Provider pool management: primary / fallback pool routing."""
 
+import random
 import structlog
 from typing import Optional
 
@@ -42,6 +43,7 @@ class PoolManager:
             if pool:
                 break
 
+        random.shuffle(backends)
         return backends
 
     def get_any_healthy_backends(self, pool: Optional[str] = None) -> list[Backend]:
@@ -55,6 +57,7 @@ class PoolManager:
                     backends.append(b)
             if pool:
                 break
+        random.shuffle(backends)
         return backends
 
     def get_pool_status(self) -> dict:

server.py

@@ -479,9 +479,24 @@ async def dashboard_sse() -> StreamingResponse:
 
 @app.get("/api/admin/backends")
 async def admin_list_backends(pool: Optional[str] = None) -> list[dict]:
-    """List all backends with masked keys (public read)."""
+    """List all backends with masked keys (public read) and real-time stats."""
     backends = list_backends(pool=pool, decrypt_key=True)
-    return [b.to_dict(mask_key=True) for b in backends]
+    result = []
+    for b in backends:
+        d = b.to_dict(mask_key=True)
+        # Inject real-time rate limiter stats
+        rl_status = rate_limiter.get_status(b.id) if rate_limiter else None
+        if rl_status:
+            utilization = rl_status.get("utilization", 0.0)
+            rpm_current = round(b.rpm_limit * utilization, 1)
+        else:
+            rpm_current = 0
+        d["stats"] = {
+            "rpm_current": rpm_current,
+            "model_count": len(b.model_mappings),
+        }
+        result.append(d)
+    return result
 
 
 @app.get("/api/admin/backends/{backend_id}")