[BUG] Reference system for RMSD CV requries same number of atoms as the system

[lohedges]

Currently the "reference" system used to provide input coordinates for the RMSD CV requires the same number of atoms as the passed system. In the tutorial, the reference uses an input PDB without hydrogens. It would be good to offer the same flexibility. This could either be handled via an heavy_only or ignore_hydrogens kwarg, or improving the flexibility of the input for the reference, e.g.:

• If it's a system and has the same number of atoms as the system, then map the selection one to one.
• If it is a system and has the same number of atoms as the selection, then map the atoms with an offset.
• Allow a user to pass in a vector of coordinates for the reference, asserting that it has the same number of entries as the selection. That way a user could load a file, get the coordinates, then use those.

For maximum flexibility, I think it should be possible to support all three of the above.

[lohedges]

Actually, this is harder than expected due to the the decoupling of alignment and RMSD, plus the new ability to handle multi-molecule input. Might just update the tutorial to use a protonated input.

[lohedges]

For now we've updated the reference file in the tutorial. See here for details.