Create setup.py file

[danielseaton]

Eventually necessary to upload the package to pip

Also useful so that we can run:

pip install -e .

from the main directory and then call the functions from notebooks located anywhere

[danielseaton]

needs to force numpy compiled against mkl (depends on installing numpy using conda? Not the default pip-installed numpy) Currently unclear if it's possible to do like this with setuptools.

[danielseaton]

May be possible to force optimised librarys with pip-installable intel packages directly:

https://software.intel.com/en-us/articles/installing-the-intel-distribution-for-python-and-intel-performance-libraries-with-pip-and