FEA：add split token and generate related resource

[txy77]

1. Update split token, generate word2vec, copy_mask, token2id, load pretrained model
2. Fix some bugs in redial, inspired and tgredial model

[txy77]

1. Fix the bugs
2. Retypeset the code

[txy77]

change the name of the variable:

1. processing -> processed_
2. split_token -> split_text

[txy77]

change the name of the variable:

1. processing -> processed_
2. split_token -> split_text

[txy77]

1. Add the version number of python package gensim

[txy77]

1. change the name of variable:
   crslabtokenizer -> Tokenizer

[txy77]

change the wat of load config

[txy77]

fix the problem of build copy_mask.npy

[txy77]

FIX: Removed unnecessary word2vec

[txy77]

FIX: Complete the integration of tokenizer classes

[txy77]

FIX: problem of data type

[txy77]

Fix: add special_token_idx to tokenizer

[txy77]

FIX: conv special_token_idx

[txy77]

FIX: variable name : CRS_Tokenizer -> crs_tokenizer

[txy77]

FIX: variable name : wordembedding -> word_embedding

[txy77]

FIX : delete redundant variable : crstokennizer

[txy77]

change variable name: BaseCrsTokenize -> BaseTokenizer

[txy77]

FIX: change as_tensor function

[txy77]

FIX: seperate the word2vec & copy_mask from dictionary

[txy77]

FIX: delete npy_dict

[txy77]

FIX: delete npy_dict

[txy77]

FIX: bert_tokenize -> BertToeknizer

[txy77]

FIX: variable name
self.Tokenizer -> self.tokenizer

[txy77]

FIX: add copy_mask = None

[txy77]

list_word -> word_list
add return copy_mask