Split bindgen simplified IR from parse/export logic

[jakelishman]

While starting to write a Rust/PyO3 raw-FFI backend for the C API via the Python package, I realised that a lot of the logic was about to be duplicated from the ctypes output. Had I realised this sooner, I'd have folded this commit into ae91986¹ to avoid an immediate refactor.

I've not attempted to de-duplicate all logic that ever might occur (for example, lifting a Function is going to look mostly similar but not identical), because there's enough little tweaks between the languages to make the abstraction awkward, and we're unlikely to have enough languages to really warrant it. For example, the ctypes binding needs to know that any custom "path" name of an Enum can't actually be exported to that name, because we use Python-space true Enum classes in the ctypes bindings for input. This restriction isn't present in other languages.

AI/LLM disclosure

• I didn't use LLM tooling, or only used it privately.
• I used the following tool to help write this PR description:
• I used the following tool to generate or modify code:

Footnotes

1. ae91986: Add auto-generated ctypes bindings to C API (Add auto-generated ctypes bindings to C API #16067) ↩

[qiskit-bot]

One or more of the following people are relevant to this code:

• @Qiskit/terra-core

[coveralls]

Coverage Report for CI Build 25402523711

Coverage decreased (-0.01%) to 87.575%

Details

• Coverage decreased (-0.01%) from the base build.
• Patch coverage: 9 uncovered changes across 2 files (185 of 194 lines covered, 95.36%).
• 26 coverage regressions across 3 files.

Uncovered Changes

File	Changed	Covered	%
crates/bindgen/src/simple_ir.rs	48	41	85.42%
crates/bindgen/src/render/ctypes.rs	143	141	98.6%

Coverage Regressions

26 previously-covered lines in 3 files lost coverage.

File	Lines Losing Coverage	Coverage
crates/qasm2/src/parse.rs	12	96.21%
crates/circuit/src/parameter/symbol_expr.rs	10	73.97%
crates/qasm2/src/lex.rs	4	91.77%

---

Coverage Stats

Relevant Lines:	121863
Covered Lines:	106721
Line Coverage:	87.57%
Coverage Strength:	961580.41 hits per line

---

💛 - Coveralls

[ShellyGarion]

This PR looks quite straightforward refactor to me.
It basically moves some functionality from ctypes.rs (previously called py_ctypes.rs) to simple_ir.rs,
to reduce future code duplication.
But since I'm not an expert with cbindgen, I would be happy to see if there are any further comments.

[raynelfss]

Ok, I took my sweet time going through the way this has been re-structured and it makes sense. It looks more clean than before (although one could argue that we now have an IR to the IR), but it looks like some of it might get replaced since Jake's PR to cbindgen has merged.

Since we need a more immediate solution, let's merge this now and update ReprType and co later.

[jakelishman]

Yeah, sorry about the churn Ray: I'd hoped to slip this PR into the previous one you reviewed, but you reviewed that faster than I could finish off what I was doing to be sure that this path was what I needed for #16145.