Catch unhandled ValueError from future timestamp

[alanhamlett]

Fixes unhandled ValueError when timestamp is too far in the future:

Traceback (most recent call last):
  File "venv/lib/python3.8/site-packages/flask/app.py", line 2072, in wsgi_app
    ctx.push()
  File "venv/lib/python3.8/site-packages/flask/ctx.py", line 404, in push
    self.session = session_interface.open_session(self.app, self.request)
  File "venv/lib/python3.8/site-packages/flask/sessions.py", line 361, in open_session
    data = s.loads(val, max_age=max_age)
  File "venv/lib/python3.8/site-packages/itsdangerous/timed.py", line 203, in loads
    base64d, timestamp = signer.unsign(
  File "venv/lib/python3.8/site-packages/itsdangerous/timed.py", line 127, in unsign
    ts_dt = self.timestamp_to_datetime(ts_int)
  File "venv/lib/python3.8/site-packages/itsdangerous/timed.py", line 47, in timestamp_to_datetime
    return datetime.fromtimestamp(ts, tz=timezone.utc)
ValueError: year 5365471 is out of range

Checklist:

• Add tests that demonstrate the correct behavior of the change. Tests should fail without the change.
• Add or update relevant docs, in the docs folder and in code.
• Add an entry in CHANGES.rst summarizing the change and linking to the issue.
• Add .. versionchanged:: entries in any relevant code docs.
• Run pre-commit hooks and fix any issues.
• Run pytest and tox, no tests failed.

[alanhamlett]

@davidism ready for review. Thanks!

[davidism]

This shouldn't be ignored. It's not freezegun that fails, it's datetime. Apparently datetime on Windows raises an OSError for an invalid timestamp value. This is expected since the handling is based on the operating system: https://bugs.python.org/issue45031

[alanhamlett]

Good catch, I see you fixed it and merged. Thanks!