Maxwell solver example

[tcew]

The Maxwell solver demos:

1. TM mode 2D time-domain Maxwell's equation with homogeneous material on triangles.
2. Full 3D time-domain Maxwell's equations with homogeneous material on tetrahedra.
3. It does not currently include quads and hexes due to their additional requirements for spurious mode free simulations.
4. The notation within the surface and volume kernels directly aligns with notation from the H&W book.
5. The maxwell solver includes an L2 error estimate if activated, with consistent mass matrix based integration of the error at W&B nodes.

[codecov]

Codecov Report

Merging #90 (75ec9d0) into staging (62efa57) will increase coverage by 0.09%.
Report is 1 commits behind head on staging.
The diff coverage is 91.12%.

❗ Current head 75ec9d0 differs from pull request most recent head 3a05adf. Consider uploading reports for the commit 3a05adf to get more accurate results

@@             Coverage Diff             @@
##           staging      #90      +/-   ##
===========================================
+ Coverage    86.19%   86.28%   +0.09%     
===========================================
  Files          285      293       +8     
  Lines        20092    20306     +214     
===========================================
+ Hits         17319    17522     +203     
- Misses        2773     2784      +11     

Files	Coverage Δ
include/comm.hpp	100.00% <ø> (ø)
solvers/maxwell/maxwell.hpp	100.00% <100.00%> (ø)
solvers/maxwell/maxwellMain.cpp	100.00% <100.00%> (ø)
solvers/maxwell/src/maxwellRun.cpp	100.00% <100.00%> (ø)
solvers/maxwell/src/maxwellStep.cpp	100.00% <100.00%> (ø)
solvers/maxwell/src/maxwellSettings.cpp	97.22% <97.22%> (ø)
solvers/maxwell/src/maxwellReport.cpp	81.25% <81.25%> (ø)
solvers/maxwell/src/maxwellSetup.cpp	89.65% <89.65%> (ø)
solvers/maxwell/src/maxwellPlotFields.cpp	84.48% <84.48%> (ø)

... and 3 files with indirect coverage changes

[noelchalmers]

Looks good. Made a couple tiny changes, and theres only a couple other things I noticed.

• profilingTips.text - Do we need this?
• The default problem size in setupTet3D.rc is real chunky, even for my workstation GPU. Maybe we can reduce this so it runs reasonably quickly?

[tcew]

Looks good. Made a couple tiny changes, and theres only a couple other things I noticed.

• profilingTips.text - Do we need this?
• The default problem size in setupTet3D.rc is real chunky, even for my workstation GPU. Maybe we can reduce this so it runs reasonably quickly?

I removed the profiling tips file and reduced the size of the box mesh in setupTet3D.rc

Are the testMaxwell.py meshes small enough ?

[noelchalmers]

lgtm