gh-493: add module for Gaussian random fields

[ntessore]

Move functionality from gaussiancl into a new module glass.grf. Transformations of Gaussian random fields can be carried out on all supported array types.

This is still a draft: core functionality in glass still needs to be expressed in terms of the new module.

Closes: #493

[ntessore]

Btw, I have switched off a lot of annoying linter rules.

Moved this to a new issue: #499

[Saransh-cpp]

Thanks, @ntessore! Looks almost ready to me! Sorry for the long review 😬 but see my suggestions and questions below -

[Saransh-cpp]

Thanks, @ntessore! Some minor formatting below. I will open up issues for the remaining bits.

[paddyroddy]

I've looked through and made various comments.

My main concern is that this is a massive PR, which I really think should be split up.

There have also been design choices which should be discussed, or just done in separate PRs

This PR also undoes a lot of the documentation work done by #381. I think it would be a shame to introduce a lot of code not following that.

[paddyroddy]

Tests are broken FYI

[ntessore]

Implemented changes from review @paddyroddy and @Saransh-cpp. Tests are broken on main.

[paddyroddy]

Implemented changes from review @paddyroddy and @Saransh-cpp. Tests are broken on main.

They're working for me? And in the UI

[ntessore]

Isn't it nox itself that's erroring out, so nothing to do with this PR?

[paddyroddy]

Isn't it nox itself that's erroring out, so nothing to do with this PR?

Ah, yes, you're correct. @Saransh-cpp you are the one in support of nox. This is the same error I saw before — although I fixed it in #471. I really think we should switch to tox, as attempted here #476. Consistent bugs are just not acceptable.

[Saransh-cpp]

I am okay with switching to tox, but the bug was more of a GH Action issue this time. The original nox issue was fixed in #471, but the PR did not fix it properly and introduced another issue with caching which popped up now (I missed it in my review 😬).

[Saransh-cpp]

Thanks, @ntessore! I believe we've created appropriate issues about the conventions/rules changed in this PR, so this should be good to go. I'll wait for @paddyroddy's review before merging.

[paddyroddy]

LGTM

[ntessore]

Thanks both, and sorry for the big & complicated PR. These are accumulated changes that would have been difficult to split up again into smaller chunks.