Add troubleshooting info about marker scaling differences between operating systems.

[TimMonko]

We need to add the marker scaling issue to the Troubleshooting page, it has popped up again (see vispy/vispy#2078) for a user. This doesn't help solve the problem, but it at least makes it more findable, hopefully.

Just for the record:
Here's the root issue, in vispy:
vispy/vispy#2078
napari issue, was closed in favor of the vispy one:
napari/napari#2894

What I think we should do is add this to the trouble shooting page:
https://napari.org/stable/troubleshooting.html

Originally posted by @psobolewskiPhD in #8050

[kesterlester]

Another helpful suggestion from someone else was that a runtime issue be perhaps printed. See suggestion here: vispy/vispy#2078 (comment) . I suspect this suggestion may have helped me more than the troubleshooting page, as I didn't really know I had a problem (and so wasn't looking for solutions) until after I'd already done lots of development. In contrast, a runtime warning (like the ones that say "feature X is now deprecated and may be removed in version Y ..." are very visible to developers (and helpful, in my opinion) -- and so may have alerted me to a potential issue earlier.

Of course, that is not an argument against also having things on the troubleshooting page.

[alisterburt]

+1 on a runtime warning!

Potentially quite loud so probably best done on layer construction

Apologies if I came off rude @kesterlester - I was having a bad day

[DragaDoncila]

@TimMonko reckon you'd have time to open a PR for this in time for 0.7.0?

[TimMonko]

Good ping @DragaDoncila . I've been vaguely following vispy/vispy#2701
@aganders3 and @brisvag, assuming a Vispy release, does this issue, by default still occur in napari? Or do we need to update napari code to use the new instanced rendering

My guess is that it's just this change that @aganders3 proposes? napari/napari@main...aganders3:napari:instanced-markers

I just want to make sure we don't add to the docs something that isn't needed!

[aganders3]

That's right, we will need a hopefully small change in napari for it to take effect.

I'm not quite sure on timing for the 0.7 release but I think @brisvag is working hard to make it happen!

Though maybe a docs update is still worthwhile?

In the vispy repo I think it was determined that emitting a warning would be difficult since the final size is determined in the shader.

[brisvag]

IMO we can get this into napari easily for 0.7.0. Unfortunately, as we said in the original issue, there is no sensible way to warn users about this without being spammy, because we can't detect when the issue is actually happening. I guess we could potentially do our own calculation based on canvas size etc. I think our best bet is having these issues be easily findable so people know how to deal with it (e.g: now by switching to instanced rendering!).

As for napari, I think we can close this (and put nothing in the docstring) once we bump vispy and default to instanced.

[DragaDoncila]

@brisvag looking at napari/#8501 this may not be such an easy fix. Do we think it's worth adding something to the troubleshooting docs or are we still hoping to have this resolved by 0.7.0? If you think the additional info in troubleshooting will be very short-lived, I'm happy to remove the milestone from this issue.

[brisvag]

I think it's easy to get that in even partially (with an option) and follow up later. In that case, we can leave this open. Either way, I wouldn't put milestones on issues like this for this reason :P