Add stack-type functions for scene management

[GammaGames]

Describe the feature
Noble.handoff(...) to add to the stack, and a Noble.return() to pop it.

Describe alternatives you've considered
Noble.transition(...) could push every time, then return would be automatic. But that would add a ton of overhead compared to the current system since it wouldn't be a blind upgrade. A "scene_stack_size" setting defaulted to 0 might help.

Alternately, a Noble.swap(...) that has a return function and it only ever goes 1 deep.

What problems would this solve or help prevent, if any

This would be useful for inventory screens, hub worlds, areas with lots of enterable buildings.

This might require a NobleScene:reenter callback too, a noop

Additional context

[This area intentionally left blank]

[Mark-LaCroix]

Umm, one of the core tenets of Noble Engine is that there's only one scene active at any time. A lot of the design, and a lot of how users are meant to understand using it, is dependent on that. It's a restriction, but deliberately so.

That said, even the projects I've worked on could benefit from examples like the ones you cited. So I'm going to close this but convert it to a discussion.

Perhaps an alternative to this might be a "scene-ish" overlay screen type that lives outside the scene management system, with it's own update loop, etc. This would perhaps help clarify these use-cases, and could hold unique functions (passing data to and from the scene that calls it, pausing/suspending scene functions as needed, getting disposed when the scene changes), on top of shedding some of the callbacks that it wouldn't need, etc. Could be called NoblePanel?