feat: support column_width in xlsx format

[Birdi7]

This PR allows user to specify column width in XLXS format. For it has a default value, which is not suitable for long values in dataset. As a result, it is not convenient to look these tables. Also, it eliminates questions like #318 in future

I didn't find tests for exporting, so omitted this. Let me know if they are required

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.11%. Comparing base (de1a47a) to head (0f7e2ef).
Report is 59 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #516      +/-   ##
==========================================
+ Coverage   90.84%   91.11%   +0.27%     
==========================================
  Files          28       28              
  Lines        2664     2702      +38     
==========================================
+ Hits         2420     2462      +42     
+ Misses        244      240       -4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Birdi7]

Hello @hugovk! Sorry for ping, but I just want to ask if you have any time for review and merge this?

[hugovk]

Thanks for the ping! Please could you add some tests that use the different possible values?

[Birdi7]

@hugovk Thank you for the review! I've added the test on check the new feature, checkout test_xlsx_column_width. Would you mind review the PR, and if everything is ok merge it to master?

[hugovk]

Some general comments. And please could we have three separate test cases that call export_set with different values for column_width: None, an integer and "adaptive".

That way we see each branch of your code is running without error, and it will be reflected in the coverage.

[Birdi7]

Thank you for such usefull comment! I've found a bug with new tests. Your comments are pushed, alongside with all three tests. What do you think on it?

[Birdi7]

Thank you for your comment! I believe this significantly improves redability. Could please take a look if the PR is fine now?

[hugovk]

Thank you! A couple of minor suggestions.

[hugovk]

3.3.0 has been released, let's target 3.4.0:

Suggested change

	.. versionchanged:: 3.3.0
	.. versionchanged:: 3.4.0

[hugovk]

Maybe something like this, to be closer to existing exceptions?

And let's start assigning exception messages to a variable first, for the reasons set out at https://github.com/henryiii/flake8-errmsg

Suggested change

	raise ValueError(f"Unsupported value `{column_width}` passed to `column_width` "
	"parameter. It supports 'adaptive' or integer values")
	msg = (
	f"Invalid value for column_width: {column_width}. "
	f"Must be 'adaptive' or an integer."
	)
	raise ValueError(msg)

[rakinson]

Hi there, I would find this enhancement useful. Do you mind to take this last step to add this feature?
Thank you!

[Birdi7]

Hi! I won't be able to finish this PR, so feel free to do that!