Bump rng_salt version to v0.40.0

[andrijapau]

Context:

With v0.40.0 of Pennylane out, we are safe to bump the rng_salt version.

Benefits:

This alters the RNG salt of our test suite, ensuring that tests from last release weren't passing due to some magic salt.

Description of the change:

This involved modifying some tests but increasing their tolerances. This was discussed with the core team to ensure we aren't accidently hiding any problems.

[sc-82812]

[github-actions]

Hello. You may have forgotten to update the changelog!
Please edit doc/releases/changelog-dev.md with:

• A one-to-two sentence description of the change. You may include a small working example for new features.
• A link back to this PR.
• Your name (or GitHub username) in the contributors section.

[astralcai]

Could you also do a search of all local_salt that's currently in the test suite and remove the ones that are not necessary for the tests to pass anymore?

[andrijapau]

@astralcai, I removed the local_salt that don't need to be there. However, lots of new tests are failing as you can see from the CI.

[astralcai]

@astralcai, I removed the local_salt that don't need to be there. However, lots of new tests are failing as you can see from the CI.

I guess now is a good time to briefly investigate if these tests are failing because of an unfortunately bad seed or if they've been passing because of a magically good seed. If it's the former, you can add a local salt to make them pass.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.60%. Comparing base (9b45754) to head (672fba1).
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #6854   +/-   ##
=======================================
  Coverage   99.60%   99.60%           
=======================================
  Files         484      484           
  Lines       46305    46305           
=======================================
  Hits        46122    46122           
  Misses        183      183           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[JerryChen97]

Seems this PR scope extended a bit to proper adjustment to those stochastic tests right?

[JerryChen97]

Seems this PR scope extended a bit to proper adjustment to those stochastic tests right?

Then probably we could also get rid of some unnecessary numeric checks here? For example, https://github.com/PennyLaneAI/pennylane/blob/61dbc7145cb9b0883e3dff817399c983db178279/tests/ops/functions/test_map_wires.py, in this test_map_wires.py no numeric checks should be needed. This also occasionally comes up in stochastic fails.

[andrijapau]

Comments to go through with team.

Edit: Went through them with team on Feb 27 2025.