Log Message:
py-nltk: updated to 3.8

Version 3.8 2022-12-12

* Refactor dispersion plot
* Provide type hints for LazyCorpusLoader variables
* Throw warning when LanguageModel is initialized with incorrect vocabulary
* Fix WordNet's all_synsets() function
* Resolve TreebankWordDetokenizer inconsistency with end-of-string contractions
* Support both iso639-3 codes and BCP-47 language tags
* Avoid DeprecationWarning in Regexp tokenizer
* Fix many doctests, add doctests to CI
* Fix bool field not being read in VerbNet
* Greatly improve time efficiency of SyllableTokenizer when tokenizing numbers
* Fix encodings of Polish udhr corpus reader
* Allow TweetTokenizer to tokenize emoji flag sequences
* Prevent LazyModule from increasing the size of nltk.__dict__
* Fix CoreNLPServer non-default port issue
* Add "acion" suffix to the Spanish SnowballStemmer
* Allow loading WordNet without OMW
* Use input() in nltk.chat.chatbot() for Jupyter support
* Fix edit_distance_align() in distance.py
* Tackle performance and accuracy regression of sentence tokenizer since NLTK 3.6.6
* Add the Iota operator to semantic logic
* Resolve critical errors in WordNet app
* Resolve critical error in CHILDES Corpus
* Make WordNet information_content() accept adjective satellites
* Add "strict=True" parameter to CoreNLP
* Resolve issue with WordNet's synset_from_sense_key
* Handle WordNet synsets that were lost in mapping
* Resolve TypeError in Boxer
* Add function to retrieve WordNet synonyms
* Warn about nonexistent OMW offsets instead of raising an error
* Fix missing ic argument in res, jcn and lin similarity functions of WordNet
* Add support for the extended OMW
* Fix LC cutoff policy of text tiling
* Optimize ConditionalFreqDist.__add__ performance
* Add Markdown corpus reader