MBCn with 5days-OddWindow

[coxipi]

Pull Request Checklist:

• This PR addresses an already opened issue (for bug fixes / features)
  • This PR fixes #xyz
• (If applicable) Documentation has been added / updated (for bug fixes / features).
• (If applicable) Tests have been added.
• 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?

• Small workaround to be able to use the 5 day grouping in MBCn, compatible with odd windows.

Does this PR introduce a breaking change?

No

Other information:

This doesn't respect the logic of Grouper, but to activate this functionality, I pass a Grouper("5D", window=7). I think only time, time.season, time.month, time.dayofyear would be allowed as strings in a proper implementation ("The usual grouping name as xarray understands it. Ex: "time.month" or "time".") To allow more complex group, I guess we would need to accept Pandas object or something.

Since MBCn doesn't currently use map_groups, implementing this hack was relatively easy.

This is also thought specifically for a 365-day calendar, that 5 is a factor of 365 (5*73).

[coveralls]

coverage: 73.694% (-0.3%) from 73.962%
when pulling 46acbef on mbcn_5d
into 066c2cc on main.

[coxipi]

It's about 5 times as faster, as expected (well, I just showed the training, but you get it)

from xclim.testing import open_dataset
import xsdba 
ds = open_dataset('sdba/ahccd_1950-2013.nc')
da = xsdba.stack_variables(ds)

group = xsdba.Grouper("5D", window=7)
ADJ = xsdba.MBCn.train(ref=da,hist=da,base_kws=dict(group=group))
ADJ.ds.af_q.compute()
# >>> 8.6 s
group = xsdba.Grouper("time.dayofyear", window=31)
ADJ = xsdba.MBCn.train(ref=da,hist=da,base_kws=dict(group=group))
ADJ.ds.af_q.compute()
# >>> 41.4 s ~ 5x slower, as expected

[coxipi]

In my tests, I had also created a specific class DQM_gpies to be able to compute DQM like MBCn, and use directly the 5d7w time grouping. I could release this version, but maybe we should instead investigate in a solution to accept 5d7w grouping in map_groups. I have a feeling it's gonna be rough though.