Improve VLLM model discovery error handling

• Add comprehensive error handling in check_model_availability method
• Provide helpful error messages with actionable solutions for 404 errors
• Warn when API token is set but model discovery is disabled

tests/unit/distribution/routers/test_routing_tables.py

@@ -52,7 +52,7 @@ class InferenceImpl(Impl):
     async def should_refresh_models(self):
         return False
 
-    async def allow_listing_models(self):
+    async def enable_model_discovery(self):
         return True
 
     async def list_models(self):