gh-137165: Add non-zero-padded Windows support for datetime.strftime

[F0RRZZ]

• Issue: Enhancement Proposal: Standardized Support for Non-Zero-Padded Day and Month Formatting in datetime.strftime #137165

[F0RRZZ]

This feature was also broken on Android (x86_64)

Now working and covered by tests

[pganssle]

@F0RRZZ Given that you closed this PR, does this mean you do not want us to go foward with it? It looks like a great start. If you don't want to continue with it, do you mind if someone takes your work and runs with it?

@vstinner @F0RRZZ Any thoughts on whether this shim might work better in time.strftime instead of datetime.strftime? I think that normally the datetime-specific extensions to time.stftime have been to cover things like time zones that don't make sense in the time module, which argues in favor of it going there, but on the other hand, there are no other similar hacks in time, so that argues in favor of not complicating that function.

[vstinner]

If strftime() is enhanced, it would be nice to do it in the time module. I have no opinion on this specific change.

[mhsmith]

@F0RRZZ has now created a new PR: #139088. You don't need to do this every time you make a change: please just push a new commit to the existing PR so the discussion is all in one place. And please don't force-push if you've already received a review, as it makes it more difficult to track the changes.

[F0RRZZ]

@pganssle I’ve opened a new pull request where I continued this work: #139088
I put this in _datetimemodule.c, since similar special cases are handled in datetime.strftime, that felt logical

@mhsmith Got it, thanks! I opened a new PR because with the large number of test commits the history felt a bit too messy, so I thought it would be cleaner to start fresh, sorry for the noise