Allow barriers in overlap circuit inputs

[caleb-johnson]

Fixes #11177

Summary

This PR allows the inputs to UnitaryOverlap to contain barriers.

Details and comments

When checking the circuit to ensure all instructions are unitary operations, we should ignore barriers.

[qiskit-bot]

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

• @Cryoris
• @Qiskit/terra-core
• @ajavadia

[jakelishman]

This looks ok to me as a temporary workaround, thanks Caleb. I think a more proper solution will be that we need to unify the "can this operation be used in gates?" logic somewhere, because at the moment the same sort of check is used in various places with quite differing results. For example, circuit_to_gate will reject Barrier as well. We can look at that in a follow-up, though, because it might involve some behavioural changes.

Could you also add a test to prevent regression?

[caleb-johnson]

Could you also add a test to prevent regression?

yes, will do

[jakelishman]

My bad, I should have spotted this in the first review: could you add a "bugfix" release note about this? Other than that (and the extra blank line in the test file that I think black will complain about), this looks fine to merge to me as a bugfix, with the proviso I had about that ideally we'll unify this logic somewhere at a larger scale. That wouldn't be suitable for backport, though, so this is good.

[coveralls]

Pull Request Test Coverage Report for Build 6774835580

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.

• 2 of 2 (100.0%) changed or added relevant lines in 1 file are covered.
• 147 unchanged lines in 13 files lost coverage.
• Overall coverage increased (+0.04%) to 86.921%

---

Files with Coverage Reduction	New Missed Lines	%
qiskit/passmanager/passmanager.py	1	91.49%
qiskit/transpiler/passes/scheduling/padding/dynamical_decoupling.py	1	94.16%
qiskit/extensions/quantum_initializer/squ.py	2	84.15%
qiskit/passmanager/base_tasks.py	2	95.89%
crates/qasm2/src/lex.rs	5	91.16%
qiskit/transpiler/target.py	5	93.78%
qiskit/primitives/backend_estimator.py	6	96.19%
qiskit/transpiler/basepasses.py	8	91.58%
qiskit/transpiler/passes/layout/sabre_layout.py	9	78.03%
qiskit/passmanager/flow_controllers.py	11	66.12%

Totals
Change from base Build 6731270262:	0.04%
Covered Lines:	74329
Relevant Lines:	85513

---

💛 - Coveralls

[caleb-johnson]

My bad, I should have spotted this in the first review: could you add a "bugfix" release note about this?

Yep, was actually going to ask about this and figured you'd tell me whether this warranted one :)

[caleb-johnson]

unify this logic somewhere at a larger scale.

If you have an inclination about where it should live and which modules could use the new function (does this only affect circuit library?), I can start an issue and try to unify it in a separate PR.

[caleb-johnson]

unify this logic somewhere at a larger scale.

If you have an inclination about where it should live and which modules could use the new function (does this only affect circuit library?), I can start an issue and try to unify it in a separate PR.

Issue here

[jakelishman]

Cool, thanks Caleb!