Improve IndexPeaks for peaks workspaces with peaks from several runs

[Dennis Mikkelson]

When a peaks workspace has peaks from several different runs, one UB will typically not index all of the peaks that really should be indexed. This is most likely due to small errors in the goniometer position information. Once a UB is obtained that indexes the peaks from all runs to some extent, that UB should be optimized for each individual run in the peaks workspace, and the optimized UB should be used to actually index the peaks in that run. This will typically assign "correct" indices to more peaks in a consistent manner across all of the runs in the workspace.

[Dennis Mikkelson]

Re 5585, IndexPeaks now indexes using per run optimzied UB

When indexing peaks in a peaks workspace, the UB matrix now is optimized for the list of peaks from each run, before actually using it to index the peaks in that run. This will increase the number of peaks indexed, in particular when the peaks workspace contains peaks from multiple runs. refs #5585

Changeset: 7aa3e9ca02f1fe4f99f7300dc5c9ec7c1ffc4536

[Dennis Mikkelson]

Re 5585, IndexPeaks now indexes using per run optimzied UB

When indexing peaks in a peaks workspace, the UB matrix now is optimized for the list of peaks from each run, before actually using it to index the peaks in that run. This will increase the number of peaks indexed, in particular when the peaks workspace contains peaks from multiple runs. refs #5585

Changeset: 7aa3e9ca02f1fe4f99f7300dc5c9ec7c1ffc4536

[Vickie Lynch]

Tried IndexPeaks for Sapphire (TOPAZ_3680-2865) peaks file. Each run was indexed separately and 582 Peaks out of 601 were indexed with a tolerance of 0.15.

[Stuart Campbell]

This ticket has been transferred to github issue ​6431