[Feature Request] TensorView: Support for overlaying Segmentation Masks on 3D Tensors

[HumbleHumbert]

Is your feature request related to a problem? Please describe.
I am using Rerun to visualize medical imaging data (MRI in .nii.gz format). I log the raw MRI volume as a Tensor and visualize it using TensorView to scroll through slices.

However, I also have a brain segmentation mask (also a 3D volume) that I need to validate against the MRI. Currently, TensorView only seems to support visualizing a single tensor at a time. I cannot overlay the segmentation mask on top of the MRI slices as I would in 2D using SegmentationImage.

Describe the solution you'd like
I would like TensorView (or a similar volume slicing view) to support:

1. Multi-tensor layering: The ability to select multiple tensor entities in the Blueprint.
2. Segmentation Overlay: Specifically, the ability to overlay an integer-based tensor (mask) on a float-based tensor (image), utilizing AnnotationContext for coloring and opacity.

Describe alternatives you've considered

• Converting the 3D segmentation to a Mesh3D using marching cubes. This works for the 3D spatial view but does not allow for precise slice-by-slice validation of the segmentation boundaries against the pixel data.
• Manually "burning" the segmentation into the MRI pixel data before logging (this destroys the original data values).

Additional context
This is a standard feature in medical imaging viewers (like ITK-SNAP or 3D Slicer) and is critical for debugging 3D segmentation models.