Calculate 3D helix axis vectors in Rosetta and use to cross-check 2D helix orientations

[rhiju]

Would be good to have RiboDraw flag when when helices are changing direction (or staying straight) in the 3D structure to make sure that the RiboDraw drawing bends or stays straight too.

Not absolutely necessary, as RiboDraw advocates laying out drawings side-by-side with a Pymol session, but for huge RNAs its possible to make mistakes ...

[everyday847]

Looking at available code:

• x3dna uses Lu's earlier work SCHNAaP/SCHNArP and that code is distributed under a license that I don't recognize. It does not support inclusion in any for-profit packages.
• curves+ is in fortran but could be translated. I can't find any license at all.

We'd probably have to roll our own. This is probably not so bad because we would only have to be very approximately correct (i.e., within 45 degrees or so!).

We'd probably want rna_motif to write out the (normalized) vector per helix rather than each pairwise angle. Then RiboDraw can compute the angle between the vectors itself.