Deezymatch

What's Changed
* Iss131/disable cosine by kasra-hosseini in https://github.com/Living-with-machines/DeezyMatch/pull/135
* v1.3.4 by kasra-hosseini in https://github.com/Living-with-machines/DeezyMatch/pull/136


**Full Changelog**: https://github.com/Living-with-machines/DeezyMatch/compare/v1.3.3...v1.3.4

Support python 3.9

This version addresses the following issues (all related to candidate ranking):
- https://github.com/Living-with-machines/DeezyMatch/issues/85
- https://github.com/Living-with-machines/DeezyMatch/issues/117
- https://github.com/Living-with-machines/DeezyMatch/issues/98
- https://github.com/Living-with-machines/DeezyMatch/issues/97
- Linting
- pandas append ---> concat

- BUGFIX: one-column option for non-Latin-alphabet corpora
- Copy the string of the first column into the second column. This way, we do not need to extend the vocabulary.

* BUGFIX: the issue with n-grams (https://github.com/Living-with-machines/DeezyMatch/issues/109)
* Define word token separators in the input file (https://github.com/Living-with-machines/DeezyMatch/issues/78)
* Prefix/suffix parameter moved as part of the mode, not preprocessing, as it applied to subword tokenization
* To generate vectors, we don't need three-column inputs anymore. We can have one column (or three columns for backward compatibility)
* Add specific datasets for each DeezyMatch functionality + Edit the README file.
* `normalizeString` and `string_split` functions are reviewed
* Improve documentation
* Several tests are added

- Update requirements.txt
- Test on python 3.9
- BUGFIX: change scenario in candidate_scenario in `candidateRanker.py`