clang-tidy option in cmake an fix for cpp 14

[MFraters]

This adds an option to run the clang-tidy check when compiling. It is off by default, because it makes compiling slower and is only useful when developing (could turn it on in debug mode in the future). I also am rerunning and fixing a select set of clang-tidy checks clang-tidy now that I am using cpp14. I am not sure whether all changes have to do with c++14 or I just missed them before.

I am also considering adding a test whether there are any issues.

[codecov]

Codecov Report

Merging #293 (d515960) into master (be243ea) will not change coverage.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #293   +/-   ##
=======================================
  Coverage   99.00%   99.00%           
=======================================
  Files          74       74           
  Lines        4228     4228           
=======================================
  Hits         4186     4186           
  Misses         42       42           

Impacted Files	Coverage Δ
source/coordinate_systems/cartesian.cc	100.00% <ø> (ø)
source/coordinate_systems/interface.cc	100.00% <ø> (ø)
source/coordinate_systems/spherical.cc	100.00% <ø> (ø)
.../continental_plate_models/composition/interface.cc	100.00% <ø> (ø)
...es/continental_plate_models/composition/uniform.cc	100.00% <ø> (ø)
...tures/continental_plate_models/grains/interface.cc	100.00% <ø> (ø)
...eatures/continental_plate_models/grains/uniform.cc	100.00% <ø> (ø)
.../continental_plate_models/temperature/adiabatic.cc	94.11% <ø> (ø)
.../continental_plate_models/temperature/interface.cc	100.00% <ø> (ø)
...res/continental_plate_models/temperature/linear.cc	100.00% <ø> (ø)
... and 64 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 be243ea...d515960. Read the comment docs.

[github-actions]

Benchmark	Master	Feature	Difference (99.9% CI)
Slab interpolation simple none	1.472 ± 0.089 (s=304)	1.457 ± 0.077 (s=313)	-2.5% .. +0.5%
Slab interpolation curved simple none	1.400 ± 0.006 (s=314)	1.399 ± 0.003 (s=332)	-0.1% .. +0.0%
Spherical slab interpolation simple none	2.032 ± 0.042 (s=213)	2.034 ± 0.045 (s=232)	-0.5% .. +0.8%
Slab interpolation simple curved CMS	2.131 ± 0.017 (s=222)	2.126 ± 0.015 (s=203)	-0.5% .. +0.0%
Spherical slab interpolation simple CMS	4.463 ± 0.163 (s=108)	4.457 ± 0.182 (s=96)	-2.0% .. +1.7%
Spherical fault interpolation simple none	2.704 ± 0.047 (s=174)	2.692 ± 0.042 (s=162)	-1.0% .. +0.2%

[MFraters]

I think these changes are sensible and considerably clean up the code. I also think they best to do right after the release, so that future development for new contributors will hopefully be easier. Although I do not think will make a real git merge conflict, I will wait on merging for #289 since it is nearly almost ready to merge and it includes some header files which will be changed. I don't think I need to wait for #264, since the changes there are not close the the changes in this pull request.

In summary, this modernizes the code for cpp14 in several areas and introduces a few readability improvements. They can be automatically checked and even fixed though cmake, but I don't think I will enforce this strictly for the time being. There is a will also be a follow up pull request to clean the code up even more (remove unused headers and fix some clang warnings (I will also need to add a non-mac clang tester, they are really different in their warnings apparently))