Ikko Hamamura

I'll modify this PR after https://github.com/Qiskit/qiskit-terra/pull/7269. (So, don't merge this)

I'll deprecate PauliTable and StabilzierTable at the same time in 0.23.

IMO, I like to stay on the safe side for these rare cases, in which circuits have different `id` and `name` but same functionality. As Imamichi-san indicated in the previous...

Since this PR is not a new feature, but a follow-up of #8604 to improve efficiency, I would put it in at 0.22.

Yes, it is up for discussion whether this is the best implementation, but we need the Parameterized SparsePauliOp as a new feature for 0.22. I will do my best to...

Probably it's not a problem, since numpy in Rust can have `PyObject`s as elements.

@woodsp-ibm I thought you reported this issue. Could you review this PR?

TBH, I don't have a strong opinion to this issue. We don't have any major rules about Qiskit errors... Since this is an API breaking, I would like to change...

need to test 0.22 (or 0.22RC)

This assemble is a bottleneck and very slow and I would like to improve it. If you have other ideas, that's fine too. Anyway, regression of the most important benchmark,...