dentate full surface workflow

[jordandekraker]

The challenge here is generating inner/outer surfaces despite the DG being so thin.

[akhanf]

also for the nativesurf workflow #327, need an approach to obtain midthickness surfaces without holes too..

I'm looking into different surface reconstruction algorithms, and it seems like the Ball-pivoting algorithm is a good choice (could work generally for both dentate and hipp).. it relies on us having some surface normal estimates, which we actually are able to get quite cleanly (from gradient of the IO coords), and the algorithm inherently deals with open surfaces, and we could control the smoothness by the ball radius.

original paper:
http://mesh.brown.edu/taubin/pdfs/bernardini-etal-tvcg99.pdf

some python implementations:
https://github.com/rodschulz/BPA
https://github.com/Lotemn102/Ball-Pivoting-Algorithm

the open3d implementation looks like the most useful one:
https://www.open3d.org/docs/latest/tutorial/Advanced/surface_reconstruction.html

(as an aside, this and some of the other surface generation techniques could be useful for computing the shrink-wrap surface that we could use to obtain a better gyrification measure @ nima - don'thave your github id on hand Nima! )