sigma-algebra generated by a function

[affeldt-aist]

Motivation for this change

This PR extracts material from the PR #1391 : the notion of sigma-algebra generated
by a function.

This is used in PR #1391 and also in the draft PR #1861 (where it is
temporarily duplicated).

These two applications should demonstrate the interest of the definition.

We are proposing this PR to simplify the review process of PR #1391 and PR #1861.

PS: We have cancelled one of the CI runs because of a Rocqnavi issue.

Checklist

• added corresponding entries in CHANGELOG_UNRELEASED.md

• added corresponding documentation in the headers

Reference: How to document

Merge policy

As a rule of thumb:

• PRs with several commits that make sense individually and that
  all compile are preferentially merged into master.
• PRs with disorganized commits are very likely to be squash-rebased.

Reminder to reviewers

• Read this Checklist
• Put a milestone if possible
• Check labels

[affeldt-aist]

@shinya-katsumata can you confirm it looks reasonable?

[shinya-katsumata]

Thank you! I'll look at it and reply to you this week (I'm now flying to Sydney).

[affeldt-aist]

Thank you! I'll look at it and reply to you this week (I'm now flying to Sydney).

As you can see, this is just adding an instance of sigma-algebra.

[affeldt-aist]

Since this is used in at least two PRs and since this is a single, rather small addition, I will maybe merge in the interest of the release.

[shinya-katsumata]

Hi, this addition looks reasonable.
Regarding what lemma to add, I need to survey existing proven lemmas before saying something meaningful...

[affeldt-aist]

Hi, this addition looks reasonable. Regarding what lemma to add, I need to survey existing proven lemmas before saying something meaningful...

Thanks, that's reassuring. As for the lemmas, we will soon discover when cleaning PR #1861