relax SVN archive

The N-state model now supports the paramagnetic centre in return_data_name().

The rdc.back_calc() user function without an ID arg will back-calculate RDCs for all alignments.

The N-state model calculate() method now stored the back-calculated RDCs and PCSs and Q factors.

N-state models with no parameters are now caught prior to optimisation.

The minimise() method now simply throws a warning and returns without doing anything.

The paramagnetic centre is now part of the parameter vector for the N-state model.

Well, only when it is not fixed.

The paramagnetic centre optimisation has been merged into the func_population() target function.

Preparation for the optimisation of the paramagnetic centre in the N-state model.

The calculation of PCS unit vectors and distances has been shifted into the new
maths_fns.paramag_centre module and into the N-state model maths_fns code. The API of
maths_fns.n_state_model has been changed so that now atomic positions, the paramagnetic centre
position, the temperature, and frequency are passed in rather than the PCS vectors and constants.
The code is now more flexible making it easy to add PRE support.

Bug fix for the RDC back_calc() function - the align_id arg was being ignored.

The RDCs were being calculated for all tensors, and then assigned to the align_id key. The results
was that the back-calculated RDCs were for the last tensor, not for align_id.

Fix for the RDC back_calc() function for when N=1 in the N-state model.

The bond vector structure was not correct.

Fixes for solely specifying that the paramag centre will be optimised.