feat: include convolution for detector pixel width in z-scan protocol

[sbillinge]

Problem

As pointed out by a referee of the paper, the pixels on the detector have a finite width. We do not account for this in our z-scan protocol. It is likely that this is mostly accounted for as an "effective" slit width, $h$, but if this is not sufficient, this can be addressed more directly

Proposed solution

revisit this in the future if there is interest from the community.