Fix calc ch

[andrewwinters5000]

Resolves #257

[ranocha]

Feel free to request a review once you're finished with the implementation. GitHub notified me about this new PR. Just a minor comment: If you write something like Closes #257, Fixes #257, or Resolves #257 instead of Addresses issue #257, GitHub will link your PR to the issue and close it once the PR is merged. Thus, you might want to edit the opening post of this PR to make use of this feature.
See https://docs.github.com/en/enterprise/2.16/user/github/managing-your-work-on-github/closing-issues-using-keywords#about-issue-references for further details.

[codecov-io]

Codecov Report

Merging #353 (01b8361) into master (2209514) will decrease coverage by 0.08%.
The diff coverage is 64.70%.

@@            Coverage Diff             @@
##           master     #353      +/-   ##
==========================================
- Coverage   90.58%   90.49%   -0.09%     
==========================================
  Files          76       78       +2     
  Lines        8881     8903      +22     
==========================================
+ Hits         8045     8057      +12     
- Misses        836      846      +10     

Flag	Coverage Δ
unittests	90.49% <64.70%> (-0.09%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/Trixi.jl	100.00% <ø> (ø)
src/callbacks_step/callbacks_step.jl	100.00% <ø> (ø)
src/callbacks_step/stepsize_dg2d.jl	85.18% <ø> (-1.03%)	⬇️
src/callbacks_step/stepsize_dg3d.jl	90.00% <ø> (-0.91%)	⬇️
src/equations/ideal_glm_mhd_2d.jl	92.65% <50.00%> (-0.04%)	⬇️
src/equations/ideal_glm_mhd_3d.jl	94.14% <50.00%> (-0.03%)	⬇️
src/callbacks_step/glm_speed.jl	65.38% <65.38%> (ø)
src/callbacks_step/glm_speed_dg.jl	66.66% <66.66%> (ø)
src/auxiliary/precompile.jl	93.67% <100.00%> (+0.08%)	⬆️
... and 1 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update eaa53dd...01b8361. Read the comment docs.

[ranocha]

Thanks for working on this!

• You changed the algorithm used to compute c_h. Just to be sure: That was done on purpose and the new way is good, isn't it?
• The same fix should be implemented in 3D.

[andrewwinters5000]

* You changed the algorithm used to compute `c_h`. Just to be sure: That was done on purpose and the new way is good, isn't it?

Yes, now c_h take the fastest magnetoacoustic wave speed into account as well as the geometry of the problem (all done through the time step size)

* The same fix should be implemented in 3D.

Indeed, that is next...

[ranocha]

LGTM and can be merged after minor revisions indicated below. Thanks for working on this, @andrewwinters5000!

[sloede]

There are just two minor changes that I think would be helpful to include. Also, I think we should follow @ranocha's suggestion for a modified MHD constructor (for 3D as well of course): https://github.com/trixi-framework/Trixi.jl/pull/353/files#r528295385

[sloede]

LGTM - feel free to merge once all test pass!

Also, congratulations: You are now the second person who ever created a new callback from scratch, which makes you an expert in my eyes ;-)

[andrewwinters5000]

Also, congratulations: You are now the second person who ever created a new callback from scratch, which makes you an expert in my eyes ;-)

Tack så mycket! ;)