CODE OPTIMIZATION: Use less GPU memory

[telegraphic]

The pipeline seems to be using much more GPU RAM than seems reasonable -- figure out why this is! Am I creating multiple copies of the same array?

[texadactyl]

Something to watch as it evolves: "Cuda Python" = cuda + numba, supported by nvidia. https://developer.nvidia.com/blog/numba-python-cuda-acceleration/

[telegraphic]

Stepped plan uses much less memory -- still optimization to do, but can now fit a full 2^20 channels on the GPU no issue.