Fix error with surrogate characters in strftime

[pganssle]

This fixes the bug and updates the tests.

[pganssle]

I'm not a core dev so I can't push directly to your branch.

[izbyshev]

@pganssle Thank you for actually fixing the problem! Would you prefer to add tests and merge it separately as the PR for bpo-34481 or combine our PRs and fix both issues at the same time?

[pganssle]

Doesn't really matter to me, but thinking about it more, we should probably do it as two separate PRs (I'll wait for yours to get merged), so that I can easily update it in response to review comments.

[izbyshev]

@pganssle You also need to handle PyUnicode_AsUTF8AndSize() inside the loop (for "%Z"), otherwise, strftime may still fail:

>>> import datetime as dt
>>> tzinfo = dt.timezone(dt.timedelta(), '\ud800')
>>> t = dt.time(tzinfo=tzinfo)
>>> t.strftime("%Z")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
UnicodeEncodeError: 'utf-8' codec can't encode character '\ud800' in position 0: surrogates not allowed