fix: Correct precedence issue in NeutronSQ Averaging

[rprospero]

I got a warning about this line in NeutronSQ module. Since divide binds faster than comparison, we're taking the inverse of an enum and comparison that to another enum, which I'm pretty sure is not the intended action.

If the original code is correct, it could use some documentation to explain the logic. As I read it now, it looks like we always get the average of squares.

[trisyoungs]

Yep, that was indeed a problem. We don't tend to generate data with normalisation applied, so this would have largely gone unnoticed.