[ECmean] enable ecmean time selection

[oloapinivad]

PR description:

This addresses #168 by correctly propagating the selection at the retrieve level.

---

• Tests are included if a new feature is included.
• Changelog is updated.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 49.42%. Comparing base (63c7fc4) to head (64c7a0d).
⚠️ Report is 5 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #178      +/-   ##
==========================================
- Coverage   49.44%   49.42%   -0.02%     
==========================================
  Files          86       86              
  Lines        7326     7328       +2     
  Branches     1268     1267       -1     
==========================================
  Hits         3622     3622              
- Misses       3251     3253       +2     
  Partials      453      453              

Flag	Coverage Δ
unittests	49.42% <ø> (-0.02%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

[mnurisso]

changelog is missing but PR seems ready

[oloapinivad]

@mcadau @jhardenberg I realized it should be possible to revert the call in the reader_data call, so that now the regrid is done after the variable selection, and solve also: #159. Only issue is that there is no test for this, I am wondering if Marco can check if this working properly...

[mcadau]

Made some tests, with both 2 and 6 months of data (so still not enough to allow the diagnostic run so far).
In both cases, this is from the aqua analysis log:

Traceback (most recent call last):
  File "/work/bb1153/b383197/AQUA-diagnostics/aqua/diagnostics/ecmean/cli_ecmean.py", line 339, in <module>
    year1, year2 = time_check(data, year1, year2, logger=logger)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/work/bb1153/b383197/AQUA-diagnostics/aqua/diagnostics/ecmean/cli_ecmean.py", line 158, in time_check
    raise NotEnoughDataError("Not enough data, exiting...")
aqua.core.exceptions.NotEnoughDataError: Not enough data, exiting...

In addition, I confirm that in the current main the diagnostic wrongly runs errorless.

I also tested 12 months of data with this branch and no #159 errors occurred. So yes, you're right, and this change solves both issues.

If you agree we can merge.
Shall I proceed with operational porting of these changes?

[oloapinivad]

That's great! Thanks for testing... perhaps you can port this to the v0.19-operational (in AQUA?)? So that we can test it extensively in the next days and if everything is as we like it we can merge both!

[mcadau]

Yes no problem!

[mcadau]

Since #2761 worked fine and tests were good on the operational as well, we can merge it also for main