relax SVN archive

Added a directory of relax results for the truncated high-precision Hansen CPMG R2eff values.

This uses the base_pipe.bz2 and r2eff_pipe.bz2 files with the high-precision R2eff errors, and hence
can be compared to the NESSY and CPMGFit results.

Fix for the phi_ex dispersion parameter scaling - the scaling was in the wrong direction.

Added parameter checks to the Relax_disp.test_hansen_cpmg_data_LM63 system test.

Speed up of the system tests using Flemming Hansen's CPMG data.

The grid search increments, function tolerance and maximum number of iterations have all been made
looser.

Two class variables have been added to the dispersion auto-analysis for fast optimisation.

This includes variables for the function tolerance and maximum number of iterations, and matches
those of the model-free auto-analysis of the dauvergne_protocol module. These will be used to
speed up the test suite.

Fixes for the checks in the Relax_disp.test_exp_fit system test.

The r2eff and i0 spin data structure keys are now strings.

Fixes for the backend of the relax_disp.plot_exp_curves user function.

This code needed to be updated for the major changes in the relax_disp branch.

Fixes for the peak intensity loading wizard for the frq.set to spectrometer.frequency user function change.

Fix for the linear constraints for the 'R2eff' model.

The A and b matrices are no longer set to None, as this kills the auto-analysis or any analysis when
constraints are turned on. Now the constraints 0 <= R2eff <= 200 and I0 >= 0 are used.

Implemented a basic graph for the relax_disp.plot_disp_curves user function.

This simply plots out the nu_CPMG value or spin-lock field verses the R2eff/R1rho values from the
experiment. The graph of the back calculated R2eff/R1rho values from the model fit is still to be
added.