interface for the distance to a feature's plane

[lhy11009]

I added a simple interface that could be used by ASPECT to call the distance_point_from_curved_planes function for computing the distance to a feature's plane surface.
I have also added a test in the unit tests for testing the implementation of the interface.

[MFraters]

Thanks for making this pull request! I think this will be a useful addition. To start let's get the tester working. Could you indent the code and add world_builder/objects/distance_from_surface.h?

[github-actions]

Benchmark	Main	Feature	Difference (99.9% CI)
Slab interpolation simple none	1.252 ± 0.002 (s=364)	1.251 ± 0.002 (s=357)	-0.1% .. -0.0%
Slab interpolation curved simple none	1.192 ± 0.004 (s=372)	1.189 ± 0.005 (s=386)	-0.3% .. -0.1%
Spherical slab interpolation simple none	2.007 ± 0.026 (s=215)	2.005 ± 0.030 (s=236)	-0.6% .. +0.3%
Slab interpolation simple curved CMS	1.470 ± 0.024 (s=336)	1.467 ± 0.026 (s=279)	-0.6% .. +0.3%
Spherical slab interpolation simple CMS	1.773 ± 0.005 (s=260)	1.772 ± 0.010 (s=250)	-0.1% .. +0.1%
Spherical fault interpolation simple none	1.669 ± 0.004 (s=275)	1.665 ± 0.004 (s=267)	-0.3% .. -0.2%
Cartesian min max surface	3.592 ± 0.035 (s=117)	3.580 ± 0.042 (s=137)	-0.8% .. +0.1%
Spherical min max surface	9.656 ± 0.014 (s=46)	9.648 ± 0.020 (s=50)	-0.2% .. +0.0%

[lhy11009]

@MFraters There is some tests failing, I am not sure if I missed something (tests / test_linux_and_macos_mpi ).

[MFraters]

It took me a bit to get to this issue. The ubuntu-latest tester images where updated and I should concider fixing them to prevent these kind of unexpected problems in the future. Can you rebase?

[lhy11009]

@MFraters I have just rebased to the main branch.

[MFraters]

Great, the tester is working again. Some small comments, mostly on documentation.

The pull request looks good to me, but can you confirm that it actually does what you want it to do when building your model? Adding a picture of that here would be cool :)

[codecov]

Codecov Report

Merging #453 (8fd0518) into main (d46936c) will increase coverage by 0.03%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main     #453      +/-   ##
==========================================
+ Coverage   90.65%   90.68%   +0.03%     
==========================================
  Files          88       89       +1     
  Lines        6246     6271      +25     
==========================================
+ Hits         5662     5687      +25     
  Misses        584      584              

Impacted Files	Coverage Δ
source/world_builder/features/interface.cc	98.07% <100.00%> (+0.07%)	⬆️
source/world_builder/features/subducting_plate.cc	99.65% <100.00%> (+0.01%)	⬆️
...rce/world_builder/objects/distance_from_surface.cc	100.00% <100.00%> (ø)
source/world_builder/world.cc	100.00% <100.00%> (ø)

---

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 d46936c...8fd0518. Read the comment docs.

[lhy11009]

@MFraters I have addressed everything in the comment. I am going to implement the designated plugin in aspect with this PR. I am thinking of generalizing that a little bit as well. Let's say the goal is to place particles relative to the surface of a feature defined in the WB. I'll update the progress.

[lhy11009]

Hi Menno, I have updated the CHANGELOG.md. I also have a better figure than the one I sent to you.

I first put a line of particles in the harzburgite layer by 5 km. The interval between particles is 0.5 km and about 9 thousand particles are entrained in the harzburgite layer (8 thousand are not plotted below the subducting plate).

The interface in this PR is used in the implementation of particles shown in this following figure:

[MFraters]

That looks great, thanks for contributing it!