measurePSF would be nice if it could also take stack into fluorcal slide

[raacampbell]

i.e. feed it an image from a really thin fluoro layer.

Currently crashes because the xy fit fails since there is no bead