Shader audio playback

[n-e-l]

As seen in https://github.com/ytanimura/sound-shader, it is possible to use glsl for writing audio data.

This would allow users, to develop generative audio in shaders. This has the added benefit that these shaders can be hot-reloaded.

@CIOSAI has already implemented audio playback in #27, now instead of filling these buffers on CPU it should be possible to fill these in using compute shaders.