Algorithm required for interpolating the efficiency wavelength dependence

[rbauststfc]

An algorithm is required that can interpolate the wavelength dependence for the analyser, flipper and polariser efficiencies. This is needed because the wavelength range of the experiment measurement may be different to the calibration measurements that were used to calculate the efficiencies.

Dirk has commented on this as follows:

It is necessary to make polarisation-analysis on TOF practically feasible without wasting too much time on calibration runs. It helps to reduce propagating excessive noise throughout the data correction. One could interpolate the data to reduce uncertainties. Using the known analytical equation gives an extra layer for checking if results are sensible just looking on the extracted parameters.

We believe that the existing InterpolatingRebin algorithm may be sufficient (and is acceptable for the MVP), however it may not be appropriate long-term because it uses cubic splines. It may be necessary for us to create a more sophisticated interpolation algorithm to use instead.

This interpolation step should be a standalone step at appropriate places in the wrapper workflow algorithm.

This is required for the Polarised SANS reduction at ISIS. Requirements still need to be gathered.