Work around a change in Git's behavior

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 74.77%. Comparing base (d0cab56) to head (02e2c9e). Report is 1 commits behind head on main.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #46 +/- ## ========================================== + Coverage 74.62% 74.77% +0.14% ========================================== Files 17 17 Lines 2735 2735 Branches 604 604 ========================================== + Hits 2041 2045 +4 + Misses 528 526 -2 + Partials 166 164 -2 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[yarikoptic]

this is unrelated to

• https://github.com/dandi/backups2datalad/issues/40 ?

[jwodder]

@yarikoptic Completely unrelated. This is about timestamps emitted by Git; that issue is about timestamps emitted by Dandi Archive.

[yarikoptic]

interesting... just odd that we are going the opposite directions here whenever in

• https://github.com/dandi/dandisets/pull/384 modifying assets to have "Z" instead of +00:00

[jwodder]

@yarikoptic I opted for +00:00 here because that's the format used by Python's datetime.isoformat() method, while Z seems to be produced by Pydantic 2.x.