Auto-bump spec version to iOS18 for models with no inputs

[john-rocky]

Summary

• Models that take no inputs require Core ML specification version ≥ 9 (iOS18). Without this fix, ct.convert(...) falls back to the default mlprogram spec version 6 (iOS15) and the resulting mlpackage fails to compile at predict time with:

  Error reading protobuf spec. validator error: Empty input is only valid in specification version >= 9. This model has version 6.

• After the MIL Program is built, detect that main has no inputs and bump the target spec version up to iOS18 (with a warning). If the user explicitly requested an older target, the existing check_deployment_compatibility step at the entry point surfaces the conflict.

Test plan

• New test TestTorchExportConversionAPI::test_no_input_model_bumps_spec_version exercises the issue repro and asserts spec.specificationVersion >= ct.target.iOS18.value, then runs verify_prediction(mlmodel).
• Manually verified surrounding behavior:
  • Model with inputs: spec version stays at 6 (no bump, no warning).
  • Explicit minimum_deployment_target=ct.target.iOS18 with no input: stays at 9, no warning.
  • Explicit minimum_deployment_target=ct.target.iOS15 with no input: bumps + the existing compatibility check raises a clear error.
• pytest test_torch_export_conversion_api.py -k "image_input or default_io_names or source_dialect_metadata" regression — 3 passed.

Fixes #2578.

[TobyRoseman]

In this case I think we should fail, with a clear error message that telling the user they need to change their deployment target.

[TobyRoseman]

Let's also check the error case here, see my other comment.

[john-rocky]

Thanks — agreed, failing with a clear error is the right behavior. Reworked in 4f65a78:

Behavior change (converter.py)

• No-input + spec version < iOS18 now raises ValueError instead of bumping. Message points the user at the fix:

  Model has no inputs, which requires a deployment target of iOS18 / macOS15 or later (specification version 9); the requested target maps to specification version 6. Re-run ct.convert(...) with minimum_deployment_target=ct.target.iOS18 (or a later target).

• Dropped the now-unused _logger import.

Test change (test_torch_export_conversion_api.py, renamed test_no_input_model_requires_ios18)

• Asserts pytest.raises(ValueError, match=r"minimum_deployment_target=ct\.target\.iOS18") for the default target.
• Then re-runs with minimum_deployment_target=ct.target.iOS18 and verifies spec.specificationVersion >= ct.target.iOS18.value + verify_prediction(mlmodel) for the happy path.

Verified locally on the same toy model (forward(self): return torch.ones(5, 5)): default target raises the new error, iOS18 target converts through MIL cleanly.