Fix for date offset strings with resample loffset

[kafitzgerald]

Closes #8399.

Reverting this minor change from #7206 allows for offset aliases to again be used with the loffset argument to resample.

Hopefully this doesn't break anything from that older PR (all tests pass). It wasn't entirely clear to me why the change was made initially. Perhaps just to simplify and this was an oversight?

[welcome]

Thank you for opening this pull request! It may take us a few days to respond here, so thank you for being patient.
If you have questions, some answers may be found in our contributing guidelines.

[dcherian]

Thanks @kafitzgerald It's great to see you here!

It wasn't entirely clear to me why the change was made initially. Perhaps just to simplify and this was an oversight?

Yes probably.

Can you add a test please?

[kafitzgerald]

Sorry about that.

And it turns out covering this particular case was fairly easy as well - just an additional parameter for the test_resample_loffset_cftimeindex test to cover date offset strings. Let me know if you had something else in mind.

I did confirm it fails prior to the change and passes after.

[spencerkclark]

Thanks @kafitzgerald — this looks perfect to me! I forgot this was the way things had been implemented before (it points to the importance of this test). Could you also add a bug fix what's new entry to the documentation?

[kafitzgerald]

Will do!

[dcherian]

Great thanks @kafitzgerald . Welcome to Xarray! It's great!

[max-sixty]

Thanks @kafitzgerald ! Welcome to Xarray!

[welcome]

Congratulations on completing your first pull request! Welcome to Xarray! We are proud of you, and hope to see you again!