port unit tests about maestroex HOT 3 CLOSED

I've ported over most of the unit tests from MAESTRO, however some of the unit tests do not work / report larger errors than expected:

• test_advect reports large errors
• test_average seems to work
• test_basestate works for planar, however some of the other problems require spherical for DIM<3 (which MAESTROeX does not support) or require networks that are not supported by Microphysics (e.g. planar-invsq requires pphotcno)
• test_diffusion's errors are larger than expected
• test_eos seems to work for both gamma law and helmholtz EoS (though error in T(p,s) is a little high for helmholtz)
• test_projection: 2d periodic is ok (errors for the nodal solver are a little high, mac solver does well), 2d wall does not work (nodal solver does not converge, mac solver has relative error > 2). 3d periodic has pretty large errors, 3d wall also does not work
• test_react: some of the tests require networks that are not supported by Microphysics. Heating tests run, but segfaults in screen.F90 for some tests with burning (e.g. aprox13). I've ported over the python scripts for generating input (xinGen.py) and checking the consistency of output (recon.py).

from maestroex.

we believe that test_advect is okay. The relative error is not really a good measure here, since the density goes to 0, so we are dividing by very small numbers.

test_react should now be fixed now.

from maestroex.

test_advect fixed in d4beb4a

from maestroex.