Fix typos in FaceDivFree interior interpolation.

[cgilet]

The problem was masked in test in amrex/Tests/DivFreePatch because all three components were the same.

Summary

Additional background

Checklist

The proposed changes:

• [ x] fix a bug or incorrect behavior in AMReX
• add new capabilities to AMReX
• changes answers in the test suite to more than roundoff level
• are likely to significantly affect the results of downstream AMReX users
• include documentation in the code and/or rst files, if appropriate

[BenWibking]

This will affect us once we turn on AMR for our MHD problems. Could the test also be updated to catch the bug?

[kngott]

I don't have the math in front of me, so I can't fully confirm this is correct, but the change looks correct, based on the pattern, so I trust it. 😄

[cgilet]

@BenWibking , yes, I'm looking at updating the test too. That's part of why this is still a draft.

Do you also use the fortran interface for divergence-preserving interpolation? I haven't looked at that, or anything specific to E&M, since my use cases don't have that.

[BenWibking]

@BenWibking , yes, I'm looking at updating the test too. That's part of why this is still a draft.

Do you also use the fortran interface for divergence-preserving interpolation? I haven't looked at that, or anything specific to E&M, since my use cases don't have that.

No, only the C++ interface.