PR summary

This PR is modeled for matplotlib#26864 axes.violinplot.

Points the PR addressed:

1. Convert Dates to Ordinals:

   Because while doing statistics on dates can get strange fast... saying 'I want a violin plot comparing measurements taken on different dates' seems completely reasonable.

   Code Response: The code converts the datetime values to ordinal values (date.toordinal()) to make them compatible with the violinplot.

2. Test Positions Instead of Stats Axis:

   I am now curious if at least the positioning values (position and width) work as desired (with the latter being a timedelta).

   Code Response: The code focuses on testing positions (positions=date_ordinals) rather than the stats axis.

3. Deprioritize Fixing Units Along Stats Axis:

   My leaning is to deprioritize fixing units along the stats axis... ask that this test case is written to focus on positions instead for the time being.

   Code Response: The code doesn't explicitly address the unit conversion for statistics but prioritizes testing positions.