Fix warning due to timezone aware pandas index

[RobbieKiwi]

Closes #451

Changes proposed in this Pull Request

Previously a timezone-aware datetime index coordinate would get converted into integers before comparison, resulting in a warning that the index does not match the pandas index.
Now there is special logic to handle this case that makes sure the coord is compared correctly.
Note this will only work in the 1D case

Checklist

• Code changes are sufficiently documented; i.e. new functions contain docstrings and further explanations may be given in doc.
• Unit tests for new features were added (if applicable).
• A note for the release notes doc/release_notes.rst of the upcoming release is included.
• I consent to the release of this PR's code under the MIT license.

[RobbieKiwi]

Maybe it would be easier to disallow timezone indexes entirely. There are some other issues that can pop up (for example when adding tz-aware data to non tz aware data) that lead to an error with numpy not understanding the datetime[ns, UTC] type when it tries to join the data

[FabianHofmann]

great @RobbieKiwi ; any chance we can generalize to multi-dim case?

[RobbieKiwi]

I added a check for the multi-dimensional case.
Also I added some logic to catch the strange type error that comes out of DataArray when you try to combine data with different timezones.
Previously this would lead to a not very useful error message something like:
"Cannot multiply pandas.Series with linopy.Variable"
Now instead there is a more useful error message which specifically mentions the timezone issue