prevent xarray from changing variable types on open_dataset

[ank1m]

GitHub Issue: 253

Description

Stitchee is inadvertently changing the data type for the main_data_quality_flag variable in TEMPO data, from int16 to float32. This impacts the SAMBAH service chain for certain edge cases (multi-file scans combining with single-file scans), because CONCISE raises an error when it encounters differing data types for the same variable.

Local test steps

Checked product/main_data_quality_flag in the output using standalone stitchee and in local Harmony environment before and after adding mask_and_scale parameter in the xarray.open_dataset function:

	Type in the original granule	Type in the stitched file	Harmony request status
Before mask_and_scale	int16	float	Failure on CONCISE step
After mask_and_scale	int16	int16	Success

Overview of integration done

• Deployed stitchee in local Harmony environment
• Performed previous egde-case requests without failures

PR Acceptance Checklist

• Unit tests added/updated and passing.
• Integration testing
• CHANGELOG.md updated
• Documentation updated (if needed).

---

📚 Documentation preview 📚: https://stitchee--254.org.readthedocs.build/en/254/

[danielfromearth]

pre-commit.ci autofix

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.42%. Comparing base (fab4ad8) to head (1f52006).
Report is 1 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #254      +/-   ##
===========================================
+ Coverage    88.11%   88.42%   +0.30%     
===========================================
  Files           11       11              
  Lines          648      648              
===========================================
+ Hits           571      573       +2     
+ Misses          77       75       -2     

Flag	Coverage Δ
integration	44.90% <ø> (ø)
unittests	85.95% <ø> (+0.30%)	⬆️

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.