implement the CMA method

[zingale]

PR summary

This does a better(?) implementation of the CMA method. By working on the interface states directly, it will be in effect for the transverse flux differences and in the high-order true-SDC as well.

PR motivation

PR checklist

• [x] test suite needs to be run on this PR
• [x] 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]

closes #1603

[zingale]

tests: http://groot.astro.sunysb.edu/Castro/test-suite/gfortran/2021-03-26-002/index.html

[zingale]

reacting_converge tests with this change:

Strang:

 density                         2.794163e+18   2.044      6.777266e+17   2.554      1.154242e+17
 xmom                            6.795828e+26   2.448      1.245025e+26   2.889      1.680538e+25
 ymom                            6.795828e+26   2.448      1.245025e+26   2.889      1.680538e+25
 rho_E                           2.450825e+35   2.351      4.802854e+34   2.742      7.178876e+33
 rho_e                           2.260573e+35   2.320      4.526276e+34   2.821      6.403446e+33
 Temp                            2.237043e+21   1.691      6.926608e+20   2.482      1.239945e+20
 rho_He4                         2.878085e+18   2.020      7.095873e+17   2.529      1.229473e+17
 rho_C12                         1.697694e+17   1.950      4.392618e+16   2.232       9.35258e+15
 rho_O16                         1.687142e+14   1.660      5.337997e+13   1.957      1.375142e+13
 rho_Fe56                           279415400   2.044          67772660   2.554          11542420

simplified-SDC:

 density                         2.794951e+18   2.044      6.776944e+17   2.554      1.154243e+17
 xmom                            6.798432e+26   2.449      1.245168e+26   2.889      1.680415e+25
 ymom                            6.798432e+26   2.449      1.245168e+26   2.889      1.680415e+25
 rho_E                           2.452176e+35   2.352      4.803923e+34   2.742      7.179435e+33
 rho_e                           2.262384e+35   2.321       4.52711e+34   2.822       6.40416e+33
 Temp                            2.236308e+21   1.691       6.92676e+20   2.481      1.240441e+20
 rho_He4                         2.878726e+18   2.020      7.095514e+17   2.529      1.229444e+17
 rho_C12                         1.697096e+17   1.950      4.392694e+16   2.231      9.356683e+15
 rho_O16                         1.686238e+14   1.659      5.338321e+13   1.957      1.375377e+13
 rho_Fe56                           279494200   2.044          67769440   2.554          11542430

[zingale]

SDC-2:

 density                         2.021788e+18   2.008      5.026002e+17   2.021      1.238369e+17
 xmom                            3.708897e+26   2.061      8.890536e+25   2.029      2.178981e+25
 ymom                            3.708897e+26   2.061      8.890536e+25   2.029      2.178981e+25
 rho_E                             2.2941e+35   2.027      5.627356e+34   2.013      1.394133e+34
 rho_e                           2.046173e+35   2.022      5.036752e+34   2.012      1.248426e+34
 Temp                              1.6252e+21   2.053      3.917765e+20   2.022      9.646864e+19
 rho_He4                         1.999378e+18   2.013      4.953558e+17   2.027      1.215214e+17
 rho_C12                         1.024488e+17   2.020      2.525206e+16   2.014      6.252355e+15
 rho_O16                         1.533022e+14   1.919      4.054309e+13   1.997      1.015876e+13
 rho_Fe56                           202178700   2.008          50260020   2.021          12383690

[maxpkatz]

Do we still want to do this?