Update decode method in tokenizer

Always add gmask in token ids

Fix bug

Fix eos token in tokenizer

Fix tokenizer config saving

Add gmask token id

Update code for slim

Drop icetk dependency

Fix decode method for torch tensor

Support single integer or empty list as input to decode (#7)

Fix generate

Fix attention_mask and position_ids

Fix encode method

Fix batch input

Implement batch generation

Change padding side

Fix typo in tokenization_chatglm.py

Remove image tokens when decoding

Update tokenizer

Add chatglm-6b