use one-sided PPM stencils at reflecting BCs

[zingale]

PR summary

PR motivation

PR checklist

• test suite needs to be run on this PR
• this PR will change answers in the test suite to more than roundoff level
• all newly-added functions have docstrings as per the coding conventions
• the CHANGES file has been updated, if appropriate
• if appropriate, this change is described in the docs

[zingale]

here's a 1-d flame_wave atmosphere run via hse_convergence_general on development with reflecting BCs:

and here's the version with these new one-sided PPM stencils:

Velocity is on the y-axis and height in the atmosphere in on the x-axis.

Notice that with these new stencils, the velocity goes to 0 at the reflecting boundary as it should.

[zingale]

the 2-d axisymmetric Sedov problem NaNs with these new BCs.

[zingale]

this seems to work well for the HSE boundary, but it does not see to work with the -x boundary in axisymmetric coords. This might be because we need to take into account the geometry somehow, like this paper suggested: https://ui.adsabs.harvard.edu/abs/1993ApJS...88..589B/abstract

[zingale]

as an example, Sedov 2-d in axisymmetry crashes after a while.

[zingale]

The way to do this to match what we do with PLM (which uses the Saltzman 1994 approach) is to only use the one-sided stencil for velocity. Then also reflect the interface states at the reflecting boundary -- this will ensure that the flux through the boundary is zero (and we won't have to rely on that hack in the Riemann solvers).

[zingale]

this has been updated to work with pslope

[zingale]

I think the MHD issues are fixed in #3109

[zingale]

for some reason, changing NUM_GROW_SRC from 3 to 4 in castro_limits.H causes O(1.e-8) diffs in the DustCollapse 3d test problem.

I am restructuring things so I don't need to do this increase, but this is odd.

[zingale]

I think that this works now. I will use this for the basis of a PR that does inflow BCs for flames, addressing #687