fix: update model filename to gpu-int4 variant

lib/core/llm/model_download_service.dart

@@ -3,7 +3,7 @@ import 'package:path_provider/path_provider.dart';
 import 'package:path/path.dart' as p;
 
 class ModelDownloadService {
-  static const modelFileName = 'gemma-2b-it-cpu-int4.bin';
+  static const modelFileName = 'gemma-2b-it-gpu-int4.bin';
   static const modelSizeBytes = 1183285248; // ~1.1 GB
 
   // Gemma 2B IT CPU int4 — MediaPipe LLM Inference model