Regenerate backend model for upstream OpenAPI drift

[Copilot]

Summary

Vendored openapi.json had drifted from upstream for the Backend schema. This updates the vendored spec and regenerates the client so the Python model surface matches current API fields.

• Spec sync (openapi.json)
  • Added Backend properties:
    • supported_error_mitigations
    • supported_gates
    • supported_native_gates
• Generated client update (ionq_core/models/backend.py)
  • Regenerated Backend model now includes the new optional fields in typed attrs + to_dict / from_dict.
• Focused model coverage (tests/test_models.py)
  • Extended backend sample + assertions to verify parse and round-trip behavior for new fields.

BACKEND_SAMPLE = {
    "backend": "qpu.aria-1",
    "status": "available",
    "supported_error_mitigations": ["debiasing"],
    "supported_gates": ["x", "y", "z"],
    "supported_native_gates": ["gpi", "gpi2", "ms"],
}

Test plan

• python -m uv run pytest

---

Important

Most code in ionq_core/ is auto-generated and overwritten on regeneration.
See CONTRIBUTING.md for which files are safe to edit.

just testing

[natestemen]

LGTM, but two questions:

1. What does supported_gates mean? If the compiler is run when non-native gates are present, why can't any gate be decomposed into the native gate set?
2. Instead of opening an issue and then manually (even if partially automated with an agent), why not just have the spec drift action open the PR directly and assign us? E.g. something like this: #55. Or is something needed for you (or someone else) to be in the loop for that I'm missing?