Duncan put QA hists for dedx into the analysis object. These are large 150x150 2d hists. Given that the correlations analyses are sorted by multplicity, these seem unnecessarily duplicated.

Also, why not include these in the fluctuations analysis?

I believe I will want to move these to the QAHists object - but will ask duncan if there is additional information I need to consider. The biggest change will be to remove their dependence on the StEStructBinning class which was designed to be part of the Correlations analysis. Unless we wish to move that class into the AnalysisMaker package?