refactor(models): enhance model management with official provider marking and extended metadata

- Add OFFICIAL_PROVIDERS set to mark first-party vendors in models.dev
- Implement official provider marking function with cache compatibility
- Extend model metadata with family, context_limit, output_limit fields
- Improve frontend model selection UI with wider panel and better search
- Add dark mode support for provider logos
- Optimize scrollbar styling for model lists
- Update deployment documentation with clearer migration steps

src/api/public/models.py

@@ -251,8 +251,8 @@ def _build_gemini_list_response(
             "version": "001",
             "displayName": m.display_name,
             "description": m.description or f"Model {m.id}",
-            "inputTokenLimit": 128000,
-            "outputTokenLimit": 8192,
+            "inputTokenLimit": m.context_limit if m.context_limit is not None else 128000,
+            "outputTokenLimit": m.output_limit if m.output_limit is not None else 8192,
             "supportedGenerationMethods": ["generateContent", "countTokens"],
             "temperature": 1.0,
             "maxTemperature": 2.0,
@@ -297,8 +297,8 @@ def _build_gemini_model_response(model_info: ModelInfo) -> dict:
         "version": "001",
         "displayName": model_info.display_name,
         "description": model_info.description or f"Model {model_info.id}",
-        "inputTokenLimit": 128000,
-        "outputTokenLimit": 8192,
+        "inputTokenLimit": model_info.context_limit if model_info.context_limit is not None else 128000,
+        "outputTokenLimit": model_info.output_limit if model_info.output_limit is not None else 8192,
         "supportedGenerationMethods": ["generateContent", "countTokens"],
         "temperature": 1.0,
         "maxTemperature": 2.0,