ronchi/foucault simulation lost ability to do autocollimation simulation

[gr5]

This is a new bug introduced in 8.3.1 (and 8.3.0). Commit was this commit:
d4e87ae
Made 12/22/2025 at 20:54pm mountain time by Dale

I believe Dale said that this change was made by AI and it looks like it because the comments are not Dale's style. The function on_makePB_clicked() was completely rewritten and in the old version we had this code:

    bool oldDoNull = md->doNull;
    if (!ui->autocollimation->isChecked()){
        md->doNull = false;
    }

And then later doNull is restored to the old state.

It may be as simple as adding this back in. I'm not sure quite where to put it because the code has changed a lot. hopefully no other features got lost.

doNull seems to be used in a few functions but I'm guessing it's computeWaveFrontFromZernikes() that is important for ronchi/foucault. And that hasn't changed. So I think this is not too hard to fix.

[gr5]

Ah. doNull was examined in null_unwrapped() in the old version but that function appears to be gone. I wonder if this change is related to the other reported bug today when reading wavefront files.

[gr5]

Actually this but is worse than I thought. 8.3.1 is showing the nulled image in foucault and ronchi view. It should show the non-nulled view (as it would look with a real ronchi or foucault tester).

[gr5]

Fixed in PR #322