Better manifold

[ALfahi]

Hello

I'm currently using the onCollisionEnter callback prop on my rigid bodies in React Three Rapier. However, I've found the current manifold object to be limited in terms of the information it exposes.

One improvement that would be incredibly helpful is access to penetration depth (e.g., manifold.depth or similar). This would allow us to distinguish between objects that are merely touching versus those that are actually intersecting or penetrating each other.

This distinction is very important since it allows to distinguish e.g a book resting on top of a table rather than it intersecting it, without any finicky work arounds.

Would it be possible to expose more detailed manifold/contact data—such as the penetration depth or the raw contact points—from Rapier in these collision callbacks?