Class Reference for ShapeCasts should mention processing each physics_frame

## Godot Version 4.4.1

### Issue

When learning about ShapeCast nodes I was unsure when I get collision results.

> Immediate collision overlaps can be done with the target_position set to Vector3(0, 0, 0) and by calling force_shapecast_update() within the same physics frame.

I **guessed** that collision_results updated once per physics frame, but I wasn't sure. It looked like I wanted `force_shapecast_update`, as I'm using it for on-demand sweeps. So I poked around there:

> ● void force_shapecast_update()
>
> Updates the collision information for the shape immediately, without waiting for the next _physics_process call. Use this method, for example, when the shape or its parent has changed state.
>
> Note: Setting enabled to true is not required for this to work.

This **implies** that it updates during each `_physics_process` and the note made me **suspect** that setting `enabled` to false would stop it from updating `collision_results` every physics frame.

`enabled` says this:

> If true, collisions will be reported.

So collisions wont be _reported_, but are they still _updated?_.

I had to dig the `cpp` code to get a better idea.

Then I looked at Raycasts:

> RayCast3D calculates intersection every physics frame, and it holds the result until the next physics frame. For an immediate raycast, or if you want to configure a RayCast3D multiple times within the same physics frame, use force_raycast_update().

This was incredibly clear and explicit in what happens and I **immediately knew** when and how I would get the results.

I found `enabled` equally lacking though.

## Solution

I think the docs for ShapeCasts need to use the same wording as RayCasts.

I also think both docs need better wording for `enabled`: does it stop calculations/updates, or only reports? Are these the same thing? If so the wording should be consistent.