Plan out budget data alignment/comparison

This is a meta ticket for thinking about the ways in which the system could support the comparison of datasets which are classified using different taxonomies. This would probably involve some internal extensions (e.g. global dimensions in cubes), and some external components (e.g. an alignment UI web app, maybe even just PyBossa). 
