fix: Clean up code

[maximus12793]

• Clean up misc comments
• Fix minor linter errors
• Fix unclosed file when reading keyword
• Default dataflow_match_score scoring logic to 0 (otherwise DMS will always be 1 which is not ideal for invalid matches)

Note: The original paper proposes a default weighting of 0.1,0.1,0.4,0.4,. Happy to make a PR for this or leave as-is.

[k4black]

Great! Thanks for the contribution! :3 Could you please check and fix the tests that failed now?

Regarding default weights - I just copied it from the original repo, but your suggestion makes sense! Happy if you'll make PR (but please note the default values in README with paper ref)

[maximus12793]

Updated tests + added logging when applicable. So actually this scoring logic does seem OK due to the weighting scheme. (for standalone tests you could return none then validate but think its fine as-is).

Couldn't find the referenced test_code_x_glue_readme_examples but updated these based on my machines fp-precision.

[codecov-commenter]

Welcome to Codecov :tada:

Once merged to your default branch, Codecov will compare your coverage reports and display the results in this comment.

Thanks for integrating Codecov - We've got you covered :open_umbrella:

[k4black]

@maximus12793 Here it is - an example from readme https://github.com/microsoft/CodeXGLUE/blob/main/Code-Code/code-to-code-trans/example.png

[maximus12793]

Hmm, yea unfortunately not sure how to repro those numbers for the second test or if they used different weights, language (c_sharp?) or what. Updated to pass linter, should be good for now!