Correct Anchor 2 calibration title

[righteousgambit]

Closes #673

Summary

• label the CalibrationOperationAnchor2 entry as Calibration - Anchor2
• add a focused regression assertion
• add the unreleased changelog entry

This is a title-only correction; operation wiring and generated calibration behavior are unchanged.

Validation

Validated against develop@2ea66c10e96d88013a04441134d3d4f8e8c88510 with Python 3.11:

• regression assertion on the base: 1 failed (Anchor1 actual, Anchor2 expected)
• regression assertion with this change: 1 passed
• ruff check passed for the source and test
• git diff --check passed

Targeted command:

poetry run python -m pytest -q tests/unit/test_calibration_operation_type.py

Changelog

- Fixed: Show the correct Anchor 2 title in the calibration operation list

Real branch verification

The repository's locked Poetry environment was installed from poetry.lock. The targeted suite passed on real fork commit b4a884fc860f, whose sole parent is pinned upstream develop@2ea66c10e96d88013a04441134d3d4f8e8c88510.

poetry run python -m pytest -q tests/unit/test_calibration_operation_type.py

Result: 1 passed.

Quality gates on the real branch:

• Ruff check: passed
• Ruff format --check: passed
• git diff --check 2ea66c10e96d88013a04441134d3d4f8e8c88510..b4a884fc860f62391fb0ec3f6e788c15ebe30139: passed

[Lyrkan]

@righteousgambit I'm sorry about how it may sound but given the other PRs/issues you opened I feel obligated to ask: is there someone behind the wheel? Or is this a fully automated process that was asked to audit the repo and create fixes for every finding?

I'm not against AI-assisted contribution at all (I actually do use LLMs quite a lot as well), but I really feel that some changes really need someone to supervise the process and check things manually, ideally with actual hardware (not this one obviously but, for instance, #675 or #676).

[Lyrkan]

Not sure that change really warrants a dedicated unit test

[righteousgambit]

No offense taken, I understand the concern. I did review each issue and actually had quite a few that I didn’t report. But I also don’t intend to spam your issues/PRs and I recognize this one, in particular, is pretty facile (but pretty sure it’s correct).

I’m interested in helping in a more dedicated fashion and don’t want to go fork my own version when y’all have put in so much effort. Do the repo maintainers have guidance for contributions beyond the workflow outlined in CONTRIBUTIONS.md?

[righteousgambit]

Also, @Lyrkan if you can clarify under what conditions maintainers want to see hardware reproduction, I’m happy to provide proof with future PRs.

[SergeBakharev]

Hi @righteousgambit ,

I'm the maintainer for the Controller repo. The general expectation for all PRs is that you have at minimum execution tested the changes with your own machine before asking others to review. Evidence of that is not strictly required, but I'm sure you can appreciate the concern given that both the issue, PR, and code appear to have been written by an autonomous AI agent.

As an example the simple copy/paste error that is the subject of this PR could have been a commit on any of the other PRs (and certainly doesn't warrant a unit test). It's not user visible and causes no application defects that I can see. Doesn't mean that it shouldn't be fixed, just that I wouldn't myself made this the subject of a PR, and it made me question if a human had reviewed this or not.

If your PRs are ready for me to look over further, please mark them as ready and not draft. You might also want to join our Discord to continue interact with the wider Carvera Community: https://discord.gg/JVjYnfu7G