Remove quantum dialect components at the end of quantum-to-ion lowering

[paul0403]

Context:
We would like the ion dialect mlir to not generate any core catalyst runtime capi stubs at the moment, since the OQD runtime needs to generate openapl instead of going through the usual catalyst runtime device interface.

To aid this, we remove all quantum dialect components from the product mlir of quantum-to-ion. With this, the ion dialect mlir can be lowered to oqd stubs a bit easier.

Description of the Change:
After the existing gate-to-pulse decomposition, the parallel protocol ops are still acting on quantum.bits. We change them to act on ion.ion by:

1. Use an ion.ion operation for each qubit. This is the equivalent of the initial quantum.extract operations at the start of a circuit.
2. Change parallel protocol operation to return ion.ion types.

Benefits:
Easier generation of openapl.

Possible Drawbacks:
Ultimately, we do want to post process the results returned by OQD backend and massage them into the form in accordance with the usual runtime device interface, e.g. probs, var, expval, etc, but at the moment this is not possible and makes openapl generation harder.

[github-actions]

Hello. You may have forgotten to update the changelog!
Please edit doc/releases/changelog-dev.md on your branch 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.