Add vocabSize to modelInfo response

Frontend needs vocab size to display dynamically instead of
hardcoded "51,200 tokens". Added vocabSize to the modelInfo
object in the research attention analysis response.

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

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

backend/model_service.py

@@ -1733,7 +1733,8 @@ async def analyze_research_attention(request: Dict[str, Any], authenticated: boo
                 "numLayers": n_layers,
                 "numHeads": n_heads,
                 "modelDimension": d_model,
-                "headDim": head_dim
+                "headDim": head_dim,
+                "vocabSize": manager.model.config.vocab_size
             },
             "generationTime": generation_time,
             "numTokensGenerated": len(generated_tokens)