Making client Qwen compatible

engine.py

@@ -127,11 +127,10 @@ from huggingface_hub import InferenceClient
 import os
 from config import HF_MODEL
 
-# 🚨 UPDATE THIS LINE: Add provider="hf-inference" to bypass third-party chat routers
+#configuring client
 client = InferenceClient(
     model=HF_MODEL, 
-    token=os.environ.get("HF_TOKEN"), 
-    provider="hf-inference"
+    token=os.environ.get("HF_TOKEN")
 )
 
 def ask_llm(prompt: str, temperature: float = 0.7, max_tokens: int = 512) -> str: