Implement Geant4 to Vecgeom in-memory converter

[mrguilima]

A list of discrepancies from the comparison of volume dumps between VGDML and converted versions of solids.gdml:

• cone1
  UnplacedCone {rmin1 0.00, rmax1 2.20, rmin2 0.00, rmax2 3.30, dz 5.55, phistart 0.00, deltaphi 2.00}
  UnplacedCone {rmin1 0.00, rmax1 2.20, rmin2 0.00, rmax2 3.30, dz 5.55, phistart 0.00, deltaphi 6.28}

• trap1
  UnplacedTrapezoid {40.000mm, -0.175rad, -0.524rad, 36.000mm, 30.000mm, 15.000mm, -0.349rad, 18.000mm, 15.000mm, 7.500mm, -0.349rad}
  UnplacedTrapezoid {40.000mm, 0.175rad, 2.618rad, 36.000mm, 30.000mm, 15.000mm, -0.364rad, 18.000mm, 15.000mm, 7.500mm, -0.364rad}

• tube100
  UnplacedTube {2.00, 3.00, 5.00, 0.00, 2.00}
  UnplacedTube {2.00, 3.00, 5.00, 0.00, 6.28}

• torus1
  UnplacedTorus2 {8.00, 10.00, 30.00, 3.14, 3.49}
  UnplacedTorus2 {8.00, 10.00, 30.00, -3.14, 3.49}

• hype1
  UnplacedHype {20.00, 24.00, 20.00, 25.00, 50.00}
  UnplacedHype {20.00, 24.00, 0.35, 0.44, 50.00}

• arb8a, arb8b: a factor 10 between vtx coordinates, like 45mm in VGDML -> 4.5mm in converted

• ellcone1
  EllipticalCone {0.03, 0.04, 100.00, 30.00}
  EllipticalCone {0.25, 0.40, 100.00, 30.00}

• tetrah1: no detailed information on solid dimensions

• no details on boolean components

[sethrj]

@mrguilima Did you nail down these problems on your branch? Can you link to the corresponding VecGeom VGDML issue and close this out if so?

[mrguilima]

Sorry for not reporting the bug fixing here, but yes, all these issues have been fixed by several commits in the g4vgConv branch. plan to close it when that branch gets merged.

[sethrj]

OK, I just wanted to confirm that these are not actually bugs in the master/develop or released branches of Celeritas, nor are they bugs in VecGeom itself. This probably should've just been left as comments/notes on the work-in-progress #557 rather than as an issue.