Review image-plate attribute of MIP task

[tcompa]

While working on fractal-analytics-platform/fractal-web#553, I noticed that the images pre- and post-projection are listed with an identical "plate" attribute:

Is this the expected behavior, or should we change it for the projected images?

[jluethi]

Oh, that's an interesting question!

I can really see this being argued either way.

My initial reaction is: We should list the mip plate as 20200812-CardiomyocyteDifferentiation14-Cycle1_mip.zarr to make those distinct.

But one could argue it's really the same plate, just with a different is_3D property