Laser should take a Jones vector for the polarization

[hartytp]

Currently Laser takes a number (-1, 0, +1) to represent the polarization. This is fine for simple things but it would be more generally useful to take a Jones vector.

Since v1.1 we work in Jones vectors elsewhere, but I haven't yet ported the Laser class to use them.

We should fix #11 in the process.

Needs proper documentation in the docstring for Laser and tests written.