coercivity_unmixing_MaxUnmix multiple plots not saving parameters

[csprain1]

In coercivity_unmixing_MaxUnmix notebook, for the example applying fits to multiple coercivity spectra, the experiment_params variable only includes parameters for the last sample and not all the samples.

[Swanson-Hysell]

We should take an approach similar to the verwey estimate notebook where we build up and return a data frame with the results for multiple specimens.