Fix mws_l1b_nc reader not scaling and masking properly

[djhoese]

Also lowercases sensor name to 'mws'.

Claude did almost all the coding on this one, but all code reviewed by me.

• Closes #xxxx
• Tests added
• Fully documented
• Add your name to AUTHORS.md if not there already

[codecov]

Codecov Report

❌ Patch coverage is 98.21429% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 96.34%. Comparing base (d2ee98f) to head (cc06db5).
⚠️ Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
satpy/readers/mws_l1b.py	97.36%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3449      +/-   ##
==========================================
- Coverage   96.34%   96.34%   -0.01%     
==========================================
  Files         465      465              
  Lines       59151    59177      +26     
==========================================
+ Hits        56990    57015      +25     
- Misses       2161     2162       +1     

Flag	Coverage Δ
behaviourtests	3.58% <0.00%> (-0.01%)	⬇️
unittests	96.43% <98.21%> (-0.01%)	⬇️

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

☔ View full report in Codecov by Harness.
📢 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.

[Manny7717]

Verified locally (head cc06db5, base ce37614)

Regression-proven: the PR's rewritten test file (with PACKING_ATTRIBUTES import removed for base compatibility) yields 4 failures on base — test_sensor (MWS vs mws), test_get_dataset_get_channeldata_counts, test_get_dataset_get_channeldata_bts, test_get_global_attributes — and 24/24 pass on head. All 4 base failures trace to the intended behavior changes, not unrelated breakage.

Core bug confirmed in the old code:

• _get_dataset_channel (old) applied only valid-range masking and NEVER applied scale_factor/add_offset — brightness temperatures came back as raw packed counts. New mask_and_scale() masks fill + valid range BEFORE scaling, casts to float32, applies scale/offset, and strips the packing attrs (they no longer describe the scaled array). Correct netCDF packing semantics.
• _get_dataset_aux_data (old) only scaled when BOTH scale_factor AND add_offset existed, and did variable.attrs["missing_value"] — a hard KeyError when the attr is absent. New path is robust to missing attrs (falls back to netCDF default_fillvals and ±inf range).
• YAML: mws_lon file_key pointed at data/navigation/mws_lat (copy-paste) — fixed to mws_lon. Real data bug.
• sensor now lowercases /attr/instrument, matching the reader config sensors: [mws] and satpy's case-sensitive supports_sensor set intersection.

Edge-checked: counts path keeps integer dtype + _FillValue attr (dtype-typed fill); float path NaN-fills; fill-vs-range precedence (FillValue → _FillValue → missing_value) sensible; xr.set_options(keep_attrs=True) preserves attrs through where/astype.

No blocking issues. Clean, well-tested fix.

[djhoese]

@Manny7717 Who are you? I'm sorry if I missed some conversation about this. Did someone ask you for these AI reviews?

[mraspaud]

LGTM, thanks an lot for the simplifications!