Orso improvements

[rozyczko]

This pull request introduces comprehensive support for ORSO file handling, including both text (.ort) and binary (.orb, NeXus/HDF5) formats. It also enhances polarization channel detection, experiment import/export, and unit handling, while making several API improvements and deprecations for clarity and maintainability.

ORSO file handling and data import/export:

• Added support for reading and writing binary ORSO (.orb, NeXus/HDF5) files alongside text .ort files, with format detection based on content rather than file extension. Binary support requires the new orb extra (h5py).
• Improved ORSO export: Project.save_experiment_as_orso serializes experiments and models using the new utility functions, and Model.as_orso now outputs the ORSO model-language dictionary (slab representation), not the internal dict.
• Repeating multilayers are now preserved during import/export, using the ORSO stack syntax and resolving stacks to retain repetition counts.
• Units declared in ORSO files are now honored and converted on load (Qz in 1/nm, lengths in nm, SLDs in 1/nm^2).

Polarization and dataset handling:

• Enhanced polarization channel detection: new functions map ORSO polarization headers to internal channels, and multi-dataset files are classified per dataset.
• Added dataset_from_datagroup to build a DataSet1D from a loaded DataGroup, preserving the parsed ORSO header for provenance and avoiding redundant parsing.
• Project experiment loading now supports pre-loaded DataGroup objects and robustly handles multi-dataset files, raising on corrupt ORSO files instead of silently falling back.

Resolution and error handling:

• Resolution and error columns are read more carefully: FWHM columns are converted to sigma, missing error columns warn instead of propagating nan, and stored Pointwise resolutions remain as variances for round-tripping.

API improvements and deprecations:

• Deprecated Project.load_orso_file in favor of explicit data and model loading methods for clarity.
• Various internal refactorings to centralize ORSO logic and reduce code duplication.

[codecov]

Codecov Report

❌ Patch coverage is 95.24941% with 20 lines in your changes missing coverage. Please review.
✅ Project coverage is 94.40%. Comparing base (c869e63) to head (264195b).

Files with missing lines	Patch %	Lines
src/easyreflectometry/orso_utils.py	95.73%	13 Missing ⚠️
src/easyreflectometry/data/measurement.py	87.50%	3 Missing ⚠️
src/easyreflectometry/data/polarized.py	89.47%	2 Missing ⚠️
src/easyreflectometry/project.py	97.01%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #402      +/-   ##
===========================================
+ Coverage    93.97%   94.40%   +0.42%     
===========================================
  Files           54       54              
  Lines         5177     5500     +323     
===========================================
+ Hits          4865     5192     +327     
+ Misses         312      308       -4     

Flag	Coverage Δ
integration	38.67% <29.21%> (-0.45%)	⬇️
unittests	94.40% <95.24%> (+0.42%)	⬆️

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

Files with missing lines	Coverage Δ
src/easyreflectometry/data/__init__.py	100.00% <100.00%> (ø)
src/easyreflectometry/model/model.py	98.12% <100.00%> (+0.65%)	⬆️
...rc/easyreflectometry/model/resolution_functions.py	98.03% <ø> (ø)
src/easyreflectometry/data/polarized.py	95.27% <89.47%> (-1.16%)	⬇️
src/easyreflectometry/project.py	94.14% <97.01%> (+1.42%)	⬆️
src/easyreflectometry/data/measurement.py	92.47% <87.50%> (+0.06%)	⬆️
src/easyreflectometry/orso_utils.py	94.40% <95.73%> (+4.56%)	⬆️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

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