Averager Alignment reflection bug #589
Description
In the great refactor of Aligners to Averagers I introduced a bug where sometimes images with reflections are off by 180*. This can be seen faintly in the tutorial.
I'll push a PR shortly that fixes this. Just waiting for the tests to run here.
FWIW, the bug actually was two distinct errors that occurred during the k loop refactoring. My bad. happy to catch them.
I will also include another minor improvement to only include the inner disc in the correlation dot product, since the outer disc is presumably only noise and 👻 artifacts.