fix: detect partial stop sequences by output suffix

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: Chessing234

fastchat/utils.py

@@ -329,8 +329,8 @@ def parse_gradio_auth_creds(filename: str):
 
 
 def is_partial_stop(output: str, stop_str: str):
-    """Check whether the output contains a partial stop str."""
-    for i in range(0, min(len(output), len(stop_str))):
+    """Check whether the output ends with a partial stop str."""
+    for i in range(1, min(len(output), len(stop_str)) + 1):
         if stop_str.startswith(output[-i:]):
             return True
     return False