Add solver for Dirac

[aromanro]

Probably it won't happen, but it's a possible enhancement.

See 'How to build an Atom' and https://arxiv.org/abs/1209.1752 The later describes also how to use a non-uniform grid. It's not a big deal to derive that one, though.

The class for Runge-Kutta solvers can be taken from the ElectricField or the CompLotkaVolterra project.