Improvements

[amerkoleci]

• Update SDL to latest stable 2.28.5
• Modernize CMake usage
• Use float for clear depth, fixes #34

[MrBrixican]

We should probably be using double as glClearDepth uses double not float, and our depth buffers have 24 bits of precision, more than the 16 max we could specify via float.

[amerkoleci]

We should probably be using double as glClearDepth uses double not float, and our depth buffers have 24 bits of precision, more than the 16 max we could specify via float.

Mmm, d3d11 for example uses float

[MrBrixican]

You know what, I screwed up on the math. floats have 32 bits, not 16. Ignore what I said, floats are fine.

[amerkoleci]

You know what, I screwed up on the math. floats have 32 bits, not 16. Ignore what I said, floats are fine.

Saw that, don't worry :)

[NoelFB]

Looks good, thanks! :D