Make ad-hoc meshes compatible with ND segmentations

[frcroth]

URL of deployed dev instance (used for testing):

• https://___.webknossos.xyz

@dieknolle3333 I think backend should be completed, though I have not tested it yet.

Steps to test:

• open an annotation for a ND dataset
• create some segments
• load meshes ad hoc
• make sure all actions for single meshes, groups and multi-selects are working
• change nd-parameter(s) and make sure it's still working.
• when moving around the nth dimension, all loaded meshes should be shown.
• when removing a mesh in one nd-position, it should be gone everywhere.

TODOs:

• show only meshes of current additionalCoordinates
• fix color changing
• fix remove meshes
• fix update meshes
• fix context menu for single mesh in segment view
• fix visibility toggle
• fix segment multiselect
• fix segment context menu in viewport
• implement saga with while-true-loop that saves the previous additional coordinates
• make sure segment view adjusts mesh actions
• move getAdditionalCoordinatesString to method, e.g. in flycam_accessor
• find a nicer way to access store in mesh saga
• make sure its working for meshes without additional params
• fix yaml.lock
• fix case where segment in not present in current additional coordianate. see case where I brush new segment and then erase all pixels.
• fix download in group menus
• investiagte whether getMeshesForAdditionalCoordinates can be used in other places
• fix that some meshes arent shown, e.g. after changing coordinates
• only remove mesh in current add coord

Issues:

• fixes Support ad-hoc meshing for ND segmentations and volume annotations #7275

---

(Please delete unneeded items, merge only when none are left open)

• Updated changelog
• Removed dev-only changes like prints and application.conf edits
• Considered common edge cases

[philippotto]

Great refactoring 🎉 The code looks way more concise now :) I left a couple of comments, but mostly only small stuff and some questions for clarifications. The finish line is visible 🏁 😃

[dieknolle3333]

one last thing... the meshes are always requested for the same position, even if the additional coordinates change. I think sometimes when creating annotation, this can lead to a bug. If I create an annotation and create segments in multiple dimensions, and I load a mesh for segment A in dimension 1, I will have to click on the segment in dimension 2 before being able to load the mesh in that dimension. However, this occurs only in the beginning of creating the dataset and I wasn't able to reproduce it afterwards. It seems to be a separate bug that has to do with the segment's position, so I feel like it's a separate issue.

[philippotto]

Excellent work! The code looks very good now and testing went well, too :) 🕺 Let's ship it! Kudos for pulling this through the finish line :)

It seems to be a separate bug that has to do with the segment's position, so I feel like it's a separate issue.

Let's iron it out later then :)

[philippotto]

Sorry 🙈 I found at least one error when reloading a pre-computed mesh:

[dieknolle3333]

Good catch, thanks! This didn't occur in my test dataset (test-agglomerate-file), but with L4 I was able to reproduce it.
I think it was only a very small bug and I hope to have fixed it now.

[philippotto]

Thank you for fixing this! 😃 I pushed a tiny commit, because I noticed that two methods had an implicit any return type. Feel free to merge now 💪