Liu-Zhang part 3

[jlchan]

This PR

1. enables the Liu-Zhang limiter for P4estMesh in 2D and 3D
2. adds a Navier-Stokes version of the Sedov example with this limiter.
3. makes the projection to the admissible set more numerically stable (necessary for the CNS Sedov example)
4. adds Base.show extensions for PositivityPreservingLimiterLiuZhang

TODO:

• Base.show specialization (moved to Liu-Zhang part 3  #3100)
• Add a CNS example
• try computing the cell averages and the local volumes together (see Liu-Zhang global cell average limiter (part 1, advection) #3063 (comment)). This probably makes more of a difference for P4estMesh, so I think we should defer it for Liu-Zhang part 3  #3100). Deferred to the next PR.