feat: add client availability tracking and error handling improvements

- Introduced `IsAvailable` and `SetUnavailable` methods to clients for availability tracking.
- Integrated availability checks in client selection logic to skip unavailable clients.
- Enhanced error handling by marking clients unavailable on specific error codes (e.g., 401, 402).
- Removed redundant quota verification logs in client reordering logic.

internal/client/client.go

@@ -41,6 +41,9 @@ type ClientBase struct {
 
 	// modelRegistry is the global model registry for tracking model availability.
 	modelRegistry *registry.ModelRegistry
+
+	// unavailable tracks whether the client is unavailable
+	isAvailable bool
 }
 
 // GetRequestMutex returns the mutex used to synchronize requests for this client.