[Preprocess][Tokenizer] Improve learn subword

[sxjscience]

Description

Solve issue https://github.com/dmlc/gluon-nlp/issues/1451

• [x] Fix the customized token support in SPM
• [x] Add a new tutorial notebook

Checklist

Essentials

• [x] PR's title starts with a category (e.g. [BUGFIX], [MODEL], [TUTORIAL], [FEATURE], [DOC], etc)
• [x] Changes are complete (i.e. I finished coding on this PR)
• [x] All changes have test coverage
• [x] Code is well-documented

cc @dmlc/gluon-nlp-team

[sxjscience]

@araitats Should be able to solve your issue after this gets merged.

[codecov[bot]]

Codecov Report

Merging #1460 (982dea1) into master (8cd1300) will increase coverage by 0.01%. The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1460      +/-   ##
==========================================
+ Coverage   85.79%   85.80%   +0.01%     
==========================================
  Files          52       52              
  Lines        6855     6855              
==========================================
+ Hits         5881     5882       +1     
+ Misses        974      973       -1     

Impacted Files	Coverage Δ
src/gluonnlp/data/tokenizers/subword_nmt.py	79.43% <0.00%> (+0.93%)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 8cd1300...982dea1. Read the comment docs.

[github-actions[bot]]

The documentation website for preview: http://gluon-nlp-staging.s3-accelerate.dualstack.amazonaws.com/PR1460/improve_learn_subword/index.html