Root transform display

In coordinate systems, add an option to enable a fixed/user root transform node that applies global transformations to the brain itself. The transform chain still behaves the same, but now the world coordinate system is actually in the frame of the second transform. This means the brain geometry can be visually tilted or translated while keeping the rest of the coordinate system untransformed.

This is to address the issue where atlases assume perfect flat orientation during experiments; however, this is difficult to guarantee, so users would like to apply small tilts or translations to better represent their _in vivo_ situation. Currently, this can be addressed by adding a fixed/user transform node at the root of the chain, but this rotates the downstream probe coordinate system and keeps the brain visually flat. This will compute the correct coordinate system values, but it is not intuitive to look at.

Start by tracing the coordinate-system transform chain and how the current fixed/user root transform is applied. Compare the existing visual brain orientation with downstream probe coordinates, then define the new root option's behavior. Done means the brain can be tilted or translated while downstream coordinates remain untransformed and correct.

Rédigé par le modèle d'indexation à partir du texte de l'issue.