Also, you can just parent empty transforms (locators). A transform needs to be a one to one relationship with the gameobject. But, there are tons of other solutions for specific use cases.
It could bring the 3D toolkit closer to the functionality in the 2D toolkit with it’s ability to subdivide a sprite and extend it with borders and 9 sprites.
Unrelated. The 2d elements only have a single transform and is nothing to do the mesh manipulation for 9-slicing. For 3D it’s same only a little more complex for uvs, but pretty simple to. Look at the Mesh class for on how to control verts. (You can also do it with a shader, not as efficient)
Arowx: Multiple transforms per game object would be a very flawed solution, which would only generate confusion for most users. The correct solution would be to use child game objects if you need additional transforms in a prefab, or use the Mesh class if you want to manipulate a mesh.