Merge Sacha's fork with fiducial sp_validation

[sachaguer]

• Added the paper plot scripts
• Added the likelihood config files for the harmonic space analysis
• Updated the scripts and notebooks I modified

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[cailmdaley]

@sachaguer want me to work on the conflicts? or you got it? they seem largely mechanical

[sachaguer]

I got it but will request your review to make sure I did not break something that was updated since I last pulled the CosmoStat repo

[sachaguer]

Ready for review @cailmdaley

[cailmdaley]

Looks good overall — the bulk is additive (paper plots, harmonic-space likelihood configs, updated notebooks), and the only library-code changes (the camb→ccl backend in cosmology.py, the nside_mask 2¹²→2¹³ bump) read as intended.

CI caught one thing, though. The PR's own checks don't fire yet — its base (stub/foundation-merge) predates the new push-triggered test CI on develop — so I ran the suite by merging current develop into the branch. It fails 23 tests with KeyError: 'mnu' at cosmology.py:283:

mnu = ccl_params.get("mnu", mnu or planck_defaults["mnu"])

planck_defaults (built just above) has no mnu entry, so the fallback raises whenever mnu isn't passed explicitly. Easy fix — the value's already at module level as PLANCK18["m_nu"] = 0.06:

planck_defaults = {
    ...,
    "mnu": PLANCK18["m_nu"],   # PLANCK18 uses m_nu; this lookup uses mnu
}

Good to merge once that's in. Bringing the branch up to develop will also light up CI on the PR itself. Thanks @sachaguer!

— Claude on behalf of Cail

[sachaguer]

Done