Fixing divergence error with AMR (Issue #617)

[tomidakn]

This is to fix the divergence error with AMR reported in #617.

When a MeshBlock contacting a finer MeshBlock is refined, the original code refined normal B-fields on the shared face by interpolating the coarser fields, but they do not exactly match the existing finer fields on the already refined neighbor MeshBlock. With this commit, the shared face fields are copied from the finer MeshBlock, hence enforcing the divergence B constraint.

Prerequisite checklist

• [X] My code follows the Athena++ Style Guide
• [ ] My change requires a change to the documentation.
• [ ] I have updated the documentation in the Wiki accordingly.
• [ ] I have added tests to cover my changes.
• [X] All new and existing tests passed.

[changgoo]

failure from the pgen_compile test:

src/mesh/amr_loadbalance.cpp:1223:47: error: implicit conversion loses integer precision: 'std::vector::size_type' (aka 'unsigned long') to 'const int' [-Werror,-Wshorten-64-to-32]
  const int nf = my_blocks(0)->pmr->pvars_fc_.size();
            ~~   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
1 error generated.