Failed to move to find tip

[DominicOram]

For collection https://ispyb.diamond.ac.uk/dc/visit/mx33658-16/dcg/10470342 Hyperion should have moved x around to try and find the tip, it didn't. From the logs it looks like Hyperion thought it found a tip. We should:

• Add more logging
• Watch out for this happening again, if it does look into averaging tip detection amongst more frames

[DominicOram]

We have seen this a couple of additional times when running overnight

[DominicOram]

https://ispyb.diamond.ac.uk/samples/sid/5003238 initially found a tip at (608, 360) but then couldn;t find the loop. We could also fix this by tweaking the edge detection settings. Suggest we find some time in shutdown to leave the edge detection on with no pin then see if we can catch when it gives a false positive

[DominicOram]

Good solution will be wait until the pin tip is "stable". In which case it will be easier to do after #863

[dperl-dls]

@dperl-dls waiting on ophyd2 oav stuff

[DominicOram]

From the data taken in #996 it appears that the main issue is on the first pin detect, which is probably because we still have old values from before the edge detection was turned on. When we do #863 this may disappear

[DominicOram]

Hotfix in DiamondLightSource/dodal#242 works fine for now, let's call this done