Remove qiskit.extensions

[Cryoris]

Summary

Remove the (sort of) deprecated qiskit.extensions module for Qiskit 1.0 and remove deprecated circuit methods (such as fredkin). I'm not sure if the reno should be more detailed, here I'm referring to the 0.45 reno, but we could also repeat that information here.

A separate PR will deprecate qiskit.extensions on the 0.46 branch.

Details and comments

• The prepare_state method should've been moved to quantumcircuit.py in the deprecation PR but I overlooked it
• Also updates some comments to not use the terminology "extensions" anymore
• Misses updating the reference images for the matplotlib tests, but I can't generate them locally so I'm misusing the CI for this

[qiskit-bot]

One or more of the the following people are requested to review this:

• @Eric-Arellano
• @enavarro51
• @javabster
• @Cryoris
• @Qiskit/terra-core
• @ajavadia
• @mtreinish
• @nkanazawa1989

[jakelishman]

This largely looks sensible to me, thanks. The lint failures are all trivial unused imports, and I commented on the two docs problems.

[jakelishman]

lol, I'm betting I wrote that "a bunch"

[Cryoris]

haha I proposed to tone it down since now it's only a few gates from the standard library (whereas before it was lot of the ones in extension) 😛

[jakelishman]

Haha yeah, that's fine. I'm mostly laughing at my terminology - I think I'm the only one of us who talks like that, and almost certainly the only one who writes code like that.

[coveralls]

Pull Request Test Coverage Report for Build 7556014030

Warning: This coverage report may be inaccurate.

We've detected an issue with your CI configuration that might affect the accuracy of this pull request's coverage report.
To ensure accuracy in future PRs, please see these guidelines.
A quick fix for this PR: rebase it; your next report should be accurate.

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.08%) to 89.404%

---

Totals
Change from base Build 7552937080:	0.08%
Covered Lines:	59224
Relevant Lines:	66243

---

💛 - Coveralls

[jakelishman]

Thanks for this - the release note is much more helpful now!

I guess the MPL test of "barrier-like objects" is no longer doing as much of its job (now it only tests real barriers), but I'm not beaten up about that - I'm not convinced that a barrier is the best drawing choice for an arbitrary "directive" anyway.

[jakelishman]

Sequence[complex] is overspecified (my pet complaint that Sequence is pretty much never correct as a type hint), but leaving aside .index and .count, Statevector would actually have already been covered by a type-hint of numpy.typing.ArrayLike, which is more appropriate.

That's not important, though, I just like complaining about Sequence.

[Cryoris]

But this would be nice to rework consistently in all type hints (though numpy.typing.ArrayLike is a bit long 😛)

[jakelishman]

Haha yeah, that's fine. I'm mostly laughing at my terminology - I think I'm the only one of us who talks like that, and almost certainly the only one who writes code like that.

[jakelishman]

Actually, I'm removing this from the merge queue and placing "on hold" until the deprecations at least have a PR open, because it's not eligible for merge without them. It's ready to merge as soon as that's in place, though.

[Cryoris]

Deprecation PR is coming up 🙂

[Cryoris]

Also on hold due to Qiskit/qiskit-aer#2023.

[jakelishman]

Yeah, it's not great that our CI is configured in such a way that we didn't notice that Aer fails to import with this change in place.

[Cryoris]

Removing on hold as Aer 0.13.2 has been released!

[jakelishman]

Thanks for getting this sorted!

[jakelishman]

Thanks for clearing up the test skips. I scanned through again, and it looks like it's all back in the state that I originally approved.