Tests for DistanceToSurface

[EricMEsch]

I just read through the tests toby added in #220 to see how i could add a test checking the Geant4 distance output. I realized that I could just use the output tobys tests already produce (namely the test-confine.lh5 file created by the confinement/gen-output test) to read in and check if the Geant4 produced distance equals the one predicted by legendhpges.

I am unsure now however, if i should just add this test to those tests (as they are under the folder confinement, and this is not really confinement). But on the other hand it seems unnecessary to recreate the same output. And i am not sure how i can add the fixture dependency if they are not part of the same CMakeLists (as in if i would put them in a new folder)?

[tdixon97]

This sounds a really good idea! I am not sure what @gipert thinks but I think its better to make a different folder for this tests, since already there is quite a lot in confinement, even if it means reproducing some outputs.
For checking the distance you don't really need the so many HPGes as I used. So you could make a simpler GDML with just a single detector.

[gipert]

I think it's ok to re-run confinement/gen-output, since it's rather inexpensive. But in general I agree that we should use fixtures, especially if simulations are costly. Having things defined in multiple CMakeLists.txt should not be a problem, what matters is that they are included in the right orded.