Clarify randomness and reproducibility of CSA operations

[sanchalitorpe-source]

This PR clarifies that CSA random and block-style operations rely on Python's global random number generator and do not re-seed on each call. This explains why repeated calls with identical parameters may produce different realizations.

This is a documentation-only change and does not modify runtime behavior.

Fixes #14

[mdjurfeldt]

Dear Sanchali, I'll look at this PR, but will be able to do it first in a few days.

[sanchalitorpe-source]

Thank you very much for the update, Mikael.

No problem at all — happy to wait. Please let me know if you’d like any wording changes or clarifications.

Best regards,
Sanchali

[mdjurfeldt]

I think this is the wrong place for this comment. There are many concepts in CSA that need to be explained. How randomness behaves is one of them and it applies not only to the _elementary module but to others as well. I think this kind of explanation should be in the README.md, the tutorial and/or the (almost nonexistent) manual.

[sanchalitorpe-source]

Thank you for the feedback — that makes sense. I agree that randomness and reproducibility are general CSA concepts and not limited to the elementary module.

My intention with this PR was to address the specific confusion raised in #14, but I’m happy to move or reframe the explanation so it lives in a more appropriate place (README, tutorial, or manual).

Would you prefer that I:

• relocate this explanation and update the PR accordingly, or
• close this PR and open a new one focused on higher-level documentation?

I’m glad to follow whatever approach fits best with CSA’s documentation structure.