Use infinite planes for slicing

[carlocastoldi]

Before submitting a pull request (PR), please read the contributing guide.

Description

What is this PR

• Bug fix
• Addition of a new feature
• Other

Why is this PR needed?
fixes a problem when intersecting some planes with the atlases.
The following code crashes without this PR:

import brainrender as br
import bgheatmaps as bgh
import numpy as np

atlas = br.Atlas(atlas_name="allen_mouse_25um")
root = atlas.get_region("root", alpha=1, color=None)
major_divisions = ["Isocortex", "OLF", "HPF", "CTXsp", "STR", "PAL", "TH", "HY", "MB", "P", "MY", "CB"]
regions_meshes = atlas.get_region(*major_divisions, alpha=1, color=None)
if not isinstance(regions_meshes, list):
    regions_meshes = [regions_meshes]

for actor in (root, *regions_meshes):
    # from Render._prepare_actor()
    mtx = [[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, -1, 0], [0, 0, 0, 1]]
    actor._mesh = actor.mesh.clone()
    actor._mesh.applyTransform(mtx) # actor._mesh.apply_transform(mtx) in newer vedo
    actor._is_transformed = True

regions_meshes = [r for r in regions_meshes if r.br_class == "brain region"]

s = bgh.slicer.Slicer(6000, "frontal", 100, root)
# CRASHES
print(s.get_structures_slice_coords(regions_meshes, root))

What does this PR do?
Instead of using a mesh as a plane, this PR uses a proper infinite plane for slicing the atlas. This effectively bypasses an open issue of vtk regarding the intersection of two meshes. See: https://discourse.vtk.org/t/vtkintersectionpolydatafilter-crashing/9428
As a bonus it is conceptually a "better" way to represent the planes.
In order to do so, I had to use a vtk function by backporting a binding from vedo=2023.4.6.

References

#11

How has this PR been tested?

I ran all examples from this repository + extensive usage in my work project.
Additionally, I made sure every combination of (frontal, horizontal, sagittal) orientations and (allen_cord_20um_v1.0, allen_mouse_25um_v1.2, mpin_zfish_1um_v1.0) altases had their plane's normal vector as well as u and v (x and y in 2D projections, respectively) in the correct direction by using the modified planner.

Is this a breaking change?

No, it should keep all the previous APIs and functionalities.

Does this PR require an update to the documentation?

No

Checklist:

• The code has been tested locally
• Tests have been added to cover all new functionality (unit & integration)
• The documentation has been updated to reflect any changes
• The code has been formatted with pre-commit

[carlocastoldi]

@LeonardoLupori I knew you were interested in this fix. Albeit not being merged, if you need it in the meantime you should be able to use this solution

[adamltyson]

@carlocastoldi Thanks for this. I've closed this PR, but included your commits in #21. If you could test that PR that would be very helpful!