Fix fill value removal for non-integer FCI data

[gerritholl]

Fix removing the _FillValue attribute for non-integer FCI data. Satpy was removing the non-existing FillValue attribute instead. Retaining the _FillValue was causing problems as described in #3058.

• Closes Writing and reading NetCDF file for FCI does not retain fill value #3058
• Tests added

[ameraner]

LGTM, thanks for the fix! I wouldn't be surprised if FillValue was used in an old test file format version available when that part of code was written, and it got fixed in the meanwhile.

[gerritholl]

There are two more occurrences of FillValue in this file and 8 more in the readers for AGRI, FY4B, GHI, MERSI, and MWS...

[ameraner]

Yes, I also noticed. Those could also be fixed to _FillValue (or the according checks could be removed alltogether since the format now is populated and should never trigger them anymore...). Up to you of course if you want to include it in this PR.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.22%. Comparing base (dc6987c) to head (f4da8b6).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3060   +/-   ##
=======================================
  Coverage   96.22%   96.22%           
=======================================
  Files         398      398           
  Lines       57330    57339    +9     
=======================================
+ Hits        55165    55174    +9     
  Misses       2165     2165           

Flag	Coverage Δ
behaviourtests	3.88% <0.00%> (-0.01%)	⬇️
unittests	96.31% <100.00%> (+<0.01%)	⬆️

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[djhoese]

I suppose the other instances should be changed too, right?

[mraspaud]

@gerritholl should we merge this or fix the other cases too first?

[gerritholl]

@gerritholl should we merge this or fix the other cases too first?

@mraspaud I fixed the other FCI case. There's also FillValue in GHI, FY4B, and MWS. I'm not familiar with those instruments/satellites, so I'm not comfortable changing those, although I would expect that the first two are similar to AGRI. Maybe someone with access to MWS (test) data can check the latter.

[coveralls]

Pull Request Test Coverage Report for Build 15020393963

Details

• 10 of 10 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.001%) to 96.327%

---

Totals
Change from base Build 15005538775:	0.001%
Covered Lines:	55434
Relevant Lines:	57548

---

💛 - Coveralls