Explore options for using TF installs that are compiled for AVX or AVX2

On some machines, this message may appear after loading tensorflow:

> Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2

As stated in this [Stack Overflow thread](https://stackoverflow.com/questions/47068709/your-cpu-supports-instructions-that-this-tensorflow-binary-was-not-compiled-to-u), this is a warning that can be safely ignored.

However, as noted in the above SO thread, there might an opportunity to gain up to 300% speedup in certain operations. Just flagging this as a potential area to document / explore.