Fix: Add attn_implementation="eager" to model switch function

When switching models via /models/switch endpoint, the model was being
loaded without attn_implementation="eager", causing Devstral/Mistral
models to return None for attention weights. This broke the Research
Attention Analyzer with error:
"TypeError: object of type 'NoneType' has no len()"

Added trust_remote_code=True and attn_implementation="eager" to match
the initial model loading configuration.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

backend/model_service.py

@@ -1245,7 +1245,9 @@ async def switch_model(request: Dict[str, Any], authenticated: bool = Depends(ve
         manager.model = AutoModelForCausalLM.from_pretrained(
             manager.model_name,
             torch_dtype=torch.float16,
-            device_map="auto"
+            device_map="auto",
+            trust_remote_code=True,
+            attn_implementation="eager"  # Required for output_attentions=True
         )
 
         # Create adapter