fix(Tokenizer): fix save_pretrained error

The origional code seems not work on tokenizer save_pretrained, self.vocab_file not exist on line 137.
This fix has been verified for save_pretrained and push_to_hub.

tokenization_chatglm.py

@@ -70,6 +70,7 @@ class ChatGLMTokenizer(PreTrainedTokenizer):
         self.name = "GLMTokenizer"
 
         self.tokenizer = SPTokenizer(vocab_file)
+        self.vocab_file = vocab_file
         self.special_tokens = {
             "<bos>": self.tokenizer.bos_id,
             "<eos>": self.tokenizer.eos_id,