[staging] 🐛 Run Examples / Compare Results has (too many) fails

[jsnel]

Version information

• pyglotaran version: staging

Describe the bug

A number of attributes are missing in the (result) datasets created from staging, compared to main, resulting in a failing result comparison tests, as can be observed in a recent PR.

Last comparison results action resulting in: ================ 141 failed, 436 passed, 49 warnings in 14.14s =================

A non-exaustive list of issues (to see all, check the actions):

• missing result attribute: 'weighted_root_mean_square_error'
• a_matrix missing suffix (e.g. a_matrix_complex1)
• k_matrix missing suffix (e.g. k_matrix_reduced_complex1)
• center_dispersion_1 missing (this is a special one, because of past renaming things)
• DAS missing decay_associated_spectra_complex1
• Missing coordinate: 'component_mc1' in 'dataset1.nc', data_var 'decay_associated_spectra_mc1'

To Reproduce

Steps to reproduce the behavior:

Open a PR to staging, observe the results of the action: Run Examples / Compare Results (pull_request)

Expected behavior

A smaller number of failing tests (maybe not 0, but a lot smaller than 141).

Tiny numerical differences can be acceptable, but we need to eliminate fails because of missing attributes.

Additional context

Note that we already worked around a lot of issues in the compat module and the examples module itself, ideally we need to do less of that.

Compat modules:

• https://github.com/glotaran/pyglotaran-extras/blob/staging_support/pyglotaran_extras/compat/compat_result.py
• https://github.com/glotaran/pyglotaran-extras/blob/staging_support/pyglotaran_extras/compat/convert_result_dataset.py

pyglotaran_examples staging changes:

• https://github.com/glotaran/pyglotaran-examples/tree/staging_rewrite/pyglotaran_examples