Manual: Add section with rules of thumb on setting up simulations

[steindev]

We should provide a section in the manual naming the numeric limits one needs to take into account when setting up simulations. This section should contain:

• temporal sampling from plasma frequency
• spatial sampling from Courant–Friedrichs–Lewy condition
• Number of particles (per Debye length)
• Cold plasma instability: temperature to which the plasma will heat
  • point out issue with moving window when initializing warm plasma
• Generation of numerical Cherenkov radiation

@ComputationalRadiationPhysics/picongpu-developers Please post suggestions for more topics to cover!

[HighIander]

• Number of particles necessary to resolve the maximum number of particles
• check that there is enough vacuum in front of the target/ that the plasma expansion goes down to sub critical density
• think about density steepening, hence one has to have good enough resolution to resolve not just the initial condition but also much higher densities

[steindev]

• Longitudinal box size needs to fit laser + target in non-moving window sims (reflections from the target should not reach boundary before laser initialization finishes)
• transverse size should match integer multiple of laser wavelength in sims w/ periodic boundaries(?), generally periodic boundaries can amplify/damp specific waves due to interference
• General section to boundary conditions would be useful, I think