Investigate very wide traces

Paul reports (slack, #drp-2d on Thursday, May 16th, 2019) that when analyzing the exposure with visitid=13183 he finds:

that Here’s the width of each fiber (except the two close to each other):

```0 2.68939032411

1 2.63398771636

2 2.08473567312
3 1.86881057574
4 1.81323757688
5 1.64066318482
6 1.53396636169
7 1.53255265524
8 1.45133603564
9 1.40209867151
10 1.38059525031
11 1.31285056784
12 1.27757783073
13 1.24954811736```
Sure looks like optics to me, what with that tilt.

This would imply huge differences between the PSF on the left and right hand side of the detector. I believe that this data should have been properly aligned, even thought they had the wrong stop (F/3.3)

This ticket is to 1. investigate this claim, 2. see if this is still the case in the data taken with the F/2.8 stop .