remove dependency on reduction folder from Inelastic

[Gesner Passos]

Inelastic uses the reduction.Reducer and reduction.ReductionSteps. Both will not be available any more as #8812 will get rid of that folder.

This means that we have to find an agreement about where to put the module that contains the Reducer and ReductionSteps. #9014 put it under SANS folder, but might not like Inelastic. At least, SANS should move the ReductionSteps to another file to allow Inelastic to use it without having to load the entire isis_reduction_steps.py.

[Dan Nixon]

Initial merge of reducer funntionality into msg_reducer

Refs #9093

Changeset: 60e14468dde8e8f0bb682b9364bebc268d5e705f

[Dan Nixon]

More refactoring and PyLint fixes

Refs #9093

Changeset: 1d9ed6a882744e45f47715e2d0dae3a8b0a111ec

[Dan Nixon]

More refactoring and PyLint corections

Refs #9093

Changeset: 7fdd0c3791dfc7e32b1b671a5acbe96083dbfa66

[Dan Nixon]

Removed seperate log file, made doctring style consistant

MSG reductions rarely seemed to output to the log, use of it has been replaced with the kernel logger at information level.

Refs #9093

Changeset: 2a92909f8f6b98bffb5f56d66280a62c5eaa8e59

[Dan Nixon]

Cleaned up impots in msg_reducer

Refs #9093

Changeset: 6f4f5555278c156125af50b270d0ba3c3813d751

[Dan Nixon]

Looking at this again, I am tempted to just migrate the reduction routines to the two Python algorithms (InelasticIndirectReduction and MSGDiffractionReduction), keeping the current reducers that inherit reducer for the time being in case any users have scripts that depend on them but make them aware they are likely to be removed.

[Dan Nixon]

Unlikely to finish this in time for 3.3

[Stuart Campbell]

This ticket has been transferred to github issue ​9936