[BUGFIX] Fix TokenEmbedding serialization with emb[emb.unknown_token] != 0

[leezu]

Follow-up on #750. #750 forgot to remove one leftover assertion, preventing serialization of TokenEmbedding with emb[emb.unknown_token] != 0.

Checklist

Essentials

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

[codecov]

Codecov Report

❗ No coverage uploaded for pull request head (tokembfixes@0019148). Click here to learn what that means.
The diff coverage is n/a.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@96420aa). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master     #763   +/-   ##
=========================================
  Coverage          ?   90.59%           
=========================================
  Files             ?       64           
  Lines             ?     6112           
  Branches          ?        0           
=========================================
  Hits              ?     5537           
  Misses            ?      575           
  Partials          ?        0

Impacted Files	Coverage Δ
src/gluonnlp/embedding/token_embedding.py	92.01% <ø> (ø)

[mli]

Job PR-763/1 is complete.
Docs are uploaded to http://gluon-nlp-staging.s3-accelerate.dualstack.amazonaws.com/PR-763/1/index.html

[mli]

Job PR-763/2 is complete.
Docs are uploaded to http://gluon-nlp-staging.s3-accelerate.dualstack.amazonaws.com/PR-763/2/index.html

[mli]

Job PR-763/3 is complete.
Docs are uploaded to http://gluon-nlp-staging.s3-accelerate.dualstack.amazonaws.com/PR-763/3/index.html