Add generic V1 Fake Backends, replace use in tests

[ElePT]

Summary

(3/4) --> This is the third PR of the FakeBackends refactoring epic.

This PR introduces generic V1 fake backends to replace the use of provider-specific backends in unit tests. It's part of the plan laid out in #10954.

Details and comments

• The generic V2 fake backend class is added in: Add fake generic and modify tests #10266
• The V2 fake backend tests are updated in: Update remaining unit tests to use FakeGeneric #10918
• This PR is blocked by the two PRs listed above, as there are tests that use both V1 and V2 backends.
• The documentation must be improved to explain what is the role of these backends in the context of the deprecation.

Replacement chart

New                 Replaces                  Occurrences

Fake20QV1           FakeAlmaden               1
                    FakeJohannesburg          1
                    FakeMelbourne             4
                    FakePoughkeepsie          1
                    FakeRueschlikon           2
                    FakeSingapore             1
                    FakeTokyo                 1

Fake27QV1Pulse      FakeHanoi                 5
                    FakeKolkata               1
                    FakeMontreal              1
                    FakeMumbai                2
                    FakeParis                 1
                    FakeToronto               1

Fake5QV1            FakeBelem                 1
                    FakeLima                  1
                    FakeLondon                1
                    FakeOurense               2
                    FakeTenerife              3
                    FakeVigo                  2
                    FakeYorktown              4

Fake65QV1Pulse      FakeAthens                1
                    FakeManhattan             1
                    FakeParis                 3
                    FakeTokyo                 1

Fake7QV1Pulse       FakeArmonk                1
                    FakeAthens                1
                    FakeNairobi               3
                    FakeOurense               2

------------------------------------------------------

Non-V1-Replacements

FakeBackendSimple   FakeArmonk                1

FakeGeneric         FakeRueschlikon           1
                    FakeTenerife              1
                    FakeTokyo                 1

cmap                FakeAlmaden               1
                    FakeMelbourne             1
                    FakeRueschlikon           3
                    FakeTenerife              2
                    FakeTokyo                 1

TBD                 FakeMumbaiFractionalCX    2

Note that some FakeBackendV1 occurrences have been replaced directly by their coupling map or the V2 FakeGeneric backend. The FakeMumbaiFractionalCX is a mumbai-inspired test-specific backend and is not part of the deprecation/migration to provider plan. I think that we could either leave it as-is or rename it to something more general like "FakeFractionalCX".

Visualization tests (joint V1 and V2 changes): --> test/visual will be left for a follow-up

• test/python/visualization/test_gate_map.py
• test/visual/mpl/circuit/test_circuit_matplotlib_drawer.py
• test/visual/mpl/graph/test_graph_matplotlib_drawer.py

Files not covered by this PR that use fake backends V1 -> to be addressed in deprecation/removal PR:

• test/python/mock/test_mock_module_deprecation.py --> will be removed in Remove qiskit.test #10998
• test/python/providers/fake_provider/test_fake_backends.py -> migrated to qiskit_ibm_runtime, Remove provider-specific fake backends,FakeProvider class and related tools in 1.0 #11376
• test/python/providers/test_fake_backends.py -> partially migrated to qiskit_ibm_runtime, Remove provider-specific fake backends,FakeProvider class and related tools in 1.0 #11376

Files not covered by this PR that use fake provider -> to be addressed in removal PR: #11376

• test/python/providers/test_backendconfiguration.py
• test/python/tools/monitor/test_backend_monitor.py
• test/python/providers/test_backendproperties.py

[qiskit-bot]

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

• @enavarro51
• @Qiskit/terra-core
• @ajavadia
• @ikkoham
• @levbishop
• @mtreinish
• @nkanazawa1989
• @t-imamichi

[coveralls]

Pull Request Test Coverage Report for Build 7733966281

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 67 (100.0%) changed or added relevant lines in 13 files are covered.
• 118 unchanged lines in 12 files lost coverage.
• Overall coverage decreased (-0.03%) to 89.452%

---

Files with Coverage Reduction	New Missed Lines	%
qiskit/transpiler/passes/layout/vf2_utils.py	1	96.34%
qiskit/primitives/containers/estimator_pub.py	2	97.53%
qiskit/primitives/containers/sampler_pub.py	2	97.01%
qiskit/transpiler/passmanager_config.py	2	95.89%
qiskit/primitives/statevector_estimator.py	3	94.55%
qiskit/init.py	4	90.48%
qiskit/transpiler/instruction_durations.py	4	90.3%
crates/qasm2/src/lex.rs	9	90.68%
qiskit/primitives/containers/bindings_array.py	9	92.37%
qiskit/primitives/containers/observables_array.py	10	90.63%

Totals
Change from base Build 7732338312:	-0.03%
Covered Lines:	59422
Relevant Lines:	66429

---

💛 - Coveralls

[ElePT]

I added the "on hold" label because this PR is built on top of #10918, but there was a lot of back and forth and a lot of conflicts, so I would like to wait until #10918 is merged to rebase and double check that I am not accidentally "undoing" any change from the previous PR here (but the essence is the same to it's still ready for review!).

[kevinhartman]

Looks mostly good to me, just a few minor things like copyright dates.

Thanks for all of this work! 😄

[kevinhartman]

What exactly makes the output different here? Is it related to how the seed is being used in the new backend?

[ElePT]

The vf2_layout pass tries to solve a subgraph isomorphism problem. From all the possible answers to this problem, it picks the one that minimizes the gate error, which depends on the backend properties. FakeTokyo and Fake20QV1 share the coupling map (because I inject it manually) but don't have the same backend properties, and this is why the optimal solution is a bit different.