Add CONDUCTOR boundary

[LudwigBoess]

This PR adds a new boundary condition - perfect conductor.

[haykh]

@LudwigBoess is this ok to merge? I struggle to find time to actually implement this CONDUCTOR on other boundaries (except for -x), so perhaps we can add in later releases if there is a demand.

[LudwigBoess]

I agree that we can do the generalization later. I cleaned up the shock pgen and edited the example .toml. So I think we can merge this now.

[LudwigBoess]

I implemented a new spatial distribution: arch::Piston that sets up the shock domain only partially filled with plasma in preparation for the moving injector.
@haykh please let me know if you're fine with the naming and implementation. As an alternative I was considering arch::PartialUniform with a more general implementation of the partially filled domain.

[LudwigBoess]

I added a first attempt at a moving injector, however it doesn't work yet

[LudwigBoess]

So, the injector is injecting and moving now. However, it produces weird stripes in the density and I can't pin-point where they come from exactly. It must have something to do with how I define xg_min and xg_max.

This is the initial condition:

This is after 116 steps:

You can test this with the shock.toml input file in the shock pgen folder.