Update condition for swapping iterators in PauliWord._matmul

[mudit2812]

Context:
We found that the wire order in Sum/Prod.terms was being swapped for the first and second term in terms that were Prods. This was due to the use of the pauli_rep. While building the pauli rep, we use PauliWord._matmul which swaps the order of self and other if len(other) >= len(self) so that we iterate through the larger PauliWord.

Example:

H = qml.prod(X(0), X(1), X(2))
Sum_coeffs, Sum_ops = H.terms()

>>> Sum_ops[0].wires, H.wires
(<Wires = [1, 0, 2]>, <Wires = [0, 1, 2]>)

Description of the Change:
Changed the condition for swapping from len(other) >= len(self) to len(other) > len(self).

Benefits:
More consistent Sum/Prod.pauli_rep, and Sum/Prod.terms() with the original operands. The above example now behaves as follows:

H = qml.prod(X(0), X(1), X(2))
Sum_coeffs, Sum_ops = H.terms()

>>> Sum_ops[0].wires, H.wires
(<Wires = [0, 1, 2]>, <Wires = [0, 1, 2]>)

Possible Drawbacks:

Related GitHub Issues:

[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.

[Qottmann]

thanks @mudit2812

Perhaps add a test ensuring that behavior

H = qml.prod(X(0), X(1), X(2))
Sum_coeffs, Sum_ops = H.terms()
assert Sum_ops[0].wires == H.wires

or update one of the existing terms() tests to also check the wires

Also changelog/bugfixes

[mudit2812]

[sc-57962]

[mudit2812]

Also changelog/bugfixes

I wouldn't call this a bug fix, as the repr wasn't incorrect, and it wasn't impacting any behaviour. I'll add tests and re-request review

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.68%. Comparing base (a8bba50) to head (288724c).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #5301      +/-   ##
==========================================
- Coverage   99.69%   99.68%   -0.01%     
==========================================
  Files         410      410              
  Lines       38230    37939     -291     
==========================================
- Hits        38113    37821     -292     
- Misses        117      118       +1     

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

[Qottmann]

Love it, thanks so much @mudit2812 ❤️

[astralcai]

Thank you for fixing this! I remember being bugged by this when we were working on the ham-tests branch. 🎸

[mlxd]

Hey folks, since we have sufficient approvals can this be merged, or is there more work coming? If more work, is there a test or two we can add to demonstrate the changes?

[mudit2812]

Hey folks, since we have sufficient approvals can this be merged, or is there more work coming? If more work, is there a test or two we can add to demonstrate the changes?

No other work @mlxd . I was just blocked by reviews on the related PR in lightning. I wanted to have that one approved before merging this so that we can push the appropriate fix to lightning ASAP after this gets merged.