Add test documenting #10561, two-digit year parsing

[jeremiahpslewis]

See issue here: #10561

[jeremiahpslewis]

@nalimilan Thank you for the feedback earlier. I believe your comments have been addressed (and the branch has been squashed per general PR requirements). Please let me know if there's anything else you would like to see in this contribution. :)

[jeremiahpslewis]

@nalimilan Given your earlier review (thanks!), just wanted to see if you had further comments after my corrections (based on @omus' feedback). :)

[jeremiahpslewis]

@omus Thanks for the updated review. I think you make a fair point, with the exception of lpad potentially not doing any work I could grok (I removed it and replaced it with string).

I think that the new test reflects the 'current behavior' of the function and is thus correct and parsimonious. However, we lose as a result the generative intuition of how dataset processing with truncated dates will easily lead to parsing mistakes, which was clearer in my original code. That said, the proper place for the intuition / operator warning is docs / a new default behavior for parsing this type of date and as such beyond the scope of this PR. :)

[omus]

I think you make a fair point, with the exception of lpad potentially not doing any work I could grok (I removed it and replaced it with string).

lpad was just adding a leading zero to 0 and 1. Definitely wasn't essential

[jeremiahpslewis]

@omus Ah, now I see it, thanks. Moved strings into the loop and converted to int in the test, think we get best of both worlds there.

[omus]

LGTM. Will merge after CI completes

[jeremiahpslewis]

Pushed a formatting change (which also triggered a new build).

[omus]

It's usually best to keep formatting changes in separate PRs but I'm okay to proceed with them here. Some minor things to change though

[jeremiahpslewis]

Exciting to see this merged, thanks for your reviews! @omus