fix: correct model type extraction for O1 model handling in litellm_a…

…i_handler.py

pr_agent/algo/ai_handlers/litellm_ai_handler.py

@@ -188,7 +188,8 @@ async def chat_completion(self, model: str, system: str, user: str, temperature:
 
             # Currently O1 does not support separate system and user prompts
             O1_MODEL_PREFIX = 'o1-'
-            if model.startswith(O1_MODEL_PREFIX):
+            model_type = model.split('/')[-1]  # 'azure/o1-' or 'o1-'
+            if model_type.startswith(O1_MODEL_PREFIX):
                 user = f"{system}\n\n\n{user}"
                 system = ""
                 get_logger().info(f"Using O1 model, combining system and user prompts")