[WIP TEST] Astyle 3.1

[MFraters]

This pull request is a test to see the differences between astye 2.04 and astyle 3.1. I am currently not thinking of switching before the next release (and maybe not even after), but I was curious to see how different it would be and provide an easy way to view and discuss the differences.

The pull request is divided into two commits. The first commit contains the persistent changes (2.04->3.1->2.04). The second commit contains the non-persistent changes.

There are not many changes (60 lines in total). A large part of the changes in both commits are in the doctest library and the rapidjson library, which maybe should not even be part of the indent (but that is a separate discussion I guess). If you exclude those libraries there are 32 lines which change within the world builder, of which 16 are persistent.

[MFraters]

Forgot to say that his came up in #420

[codecov]

Codecov Report

Merging #421 (5fa8989) into main (4d33def) will not change coverage.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main     #421   +/-   ##
=======================================
  Coverage   99.50%   99.50%           
=======================================
  Files          86       86           
  Lines        5038     5038           
=======================================
  Hits         5013     5013           
  Misses         25       25           

Impacted Files	Coverage Δ
source/parameters.cc	99.65% <ø> (ø)
source/utilities.cc	99.35% <ø> (ø)
app/main.cc	99.26% <100.00%> (ø)
source/features/fault.cc	99.63% <100.00%> (ø)
source/features/subducting_plate.cc	99.63% <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 4d33def...5fa8989. Read the comment docs.

[github-actions]

Benchmark	Main	Feature	Difference (99.9% CI)
Slab interpolation simple none	1.356 ± 0.063 (s=355)	1.355 ± 0.060 (s=311)	-1.2% .. +1.1%
Slab interpolation curved simple none	1.180 ± 0.006 (s=382)	1.179 ± 0.003 (s=383)	-0.1% .. +0.1%
Spherical slab interpolation simple none	1.951 ± 0.034 (s=217)	1.950 ± 0.034 (s=247)	-0.6% .. +0.5%
Slab interpolation simple curved CMS	1.198 ± 0.003 (s=380)	1.198 ± 0.003 (s=374)	-0.1% .. +0.0%
Spherical slab interpolation simple CMS	1.668 ± 0.010 (s=278)	1.667 ± 0.009 (s=264)	-0.2% .. +0.1%
Spherical fault interpolation simple none	1.804 ± 0.064 (s=256)	1.806 ± 0.063 (s=245)	-0.9% .. +1.2%
Cartesian min max surface	3.576 ± 0.031 (s=119)	3.584 ± 0.028 (s=135)	-0.1% .. +0.6%
Spherical min max surface	9.132 ± 0.015 (s=52)	9.132 ± 0.013 (s=49)	-0.1% .. +0.1%

[bangerth]

Yes, this really isn't very much. It would be interesting to see how that looks like for ASPECT.

I also think you're right that you should exclude imported projects from indentation. These ought to be preserved in exactly the state they were imported in.