relax SVN archive

Fixes for the free rotor frame order model system test scripts.

The average position alpha Euler angle is not defined in these models!

Added a system test script for the iso cone, free rotor frame order test data.

Added Monte Carlo simulations to all the frame order model optimisation test scripts.

Created a second iso cone, free rotor frame order model - this time with a tilted axis.

The centre of the distribution is an axis tilted 15 degrees from the pivot-CoM vector.

Fix for the pymol.cone_pdb user function for frame order models with no alpha average pos angle.

This is specifically for the isotropic cone, free rotor model. But other models might also be
affected by this fix.

Removed the back_calc.bz2 save file as it is far, far too big!

Changes for the isotropic cone, free rotor frame order test data.

The distribution of structures consists of 5000 rather than 500 models now, so the PDB file is no
longer stored. This number is needed to properly sample the space. Monte Carlo simulations have
also been added.

Fixes for the Monte Carlo simulations of the iso cone, free rotor frame order model.

The derivative cone_theta parameter (calculated from cone_s1) now is handled in the Monte Carlo
simulations and an error analysis on it is performed.

Fix for the internal structural object add_atom() method for when no models or molecules are present.

Updates for the new cone PDB file where the negative image is in a second model.