Fix: Remove unsupported use_xformers_attention parameter

app/models/huggingface_local.py

@@ -149,7 +149,6 @@ class HuggingFaceLocal(BaseLLM):
             temperature=temperature,
             top_p=top_p,
             use_cache=True,  # CRITICAL: Enable KV cache
-            use_xformers_attention=False,  # CPU doesn't support this
             eos_token_id=self.tokenizer.eos_token_id,
             pad_token_id=self.tokenizer.eos_token_id if self.tokenizer.pad_token_id is None else self.tokenizer.pad_token_id,
         )