feat(models): register Gemini 3.5 Flash with dynamic thinking levels

- Added new model `gemini-3.5-flash` to the registry with enhanced intelligence and speed capabilities.
- Supports extended thinking levels (`minimal`, `low`, `medium`, `high`) and dynamic adjustments.
- Expanded generation methods, including content creation and token counting.

internal/registry/models/models.json

@@ -762,6 +762,36 @@
       "supportedGenerationMethods": [
         "predict"
       ]
+    },
+    {
+      "id": "gemini-3.5-flash",
+      "object": "model",
+      "created": 1779235200,
+      "owned_by": "google",
+      "type": "gemini",
+      "display_name": "Gemini 3.5 Flash",
+      "name": "models/gemini-3.5-flash",
+      "version": "3.5",
+      "description": "Our most intelligent model built for speed, combining frontier intelligence with superior search and grounding.",
+      "inputTokenLimit": 1048576,
+      "outputTokenLimit": 65536,
+      "supportedGenerationMethods": [
+        "generateContent",
+        "countTokens",
+        "createCachedContent",
+        "batchGenerateContent"
+      ],
+      "thinking": {
+        "min": 128,
+        "max": 32768,
+        "dynamic_allowed": true,
+        "levels": [
+          "minimal",
+          "low",
+          "medium",
+          "high"
+        ]
+      }
     }
   ],
   "gemini-cli": [