Fix sdba Grouper.get_coordinate after xarray update

[aulemahal]

Pull Request Checklist:

• This PR addresses an already opened issue (for bug fixes / features)
  • This PR fixes adapt_freq broken with 360_day and group 31-doy #2037
• Tests for the changes have been added (for bug fixes / features)
  • (If applicable) Documentation has been added / updated (for bug fixes / features)
• CHANGELOG.rst has been updated (with summary of main changes)
  • Link to issue (:issue:number) and pull request (:pull:number) has been added

What kind of change does this PR introduce?

• Uses a dumber way of getting the max doy, that doesn't rely on a private xarray function. There is a public xarray function in >=2024.09, but we still support >=2023.11.

Does this PR introduce a breaking change?

Yes, only if a non-uniform calendar was used with a subset of years that did not include a leap year. The max doy is now static (standard => 366, for example) and not dynamic (standard, 2004 => 366).

[coveralls]

coverage: 89.962%. remained the same
when pulling f217d13 on fix-Grouper-360
into e34df04 on main.