TimeSeries demand and supply for commods as root_metrics

[nuclearkatie]

It doesn't seem like there is currently any support for the demand or supply TimeSeries tables for each commod in a simulation. They're not considered root metrics (because the table names contain the unique simulation commods).

Same goes for the AgentState tables, because they contain the library and archetype names in the table name, e.g.

evaler.eval("AgentState_cycamore_EnrichmentInfo")

throws a KeyError. There is a cycamore-specific root metric file, but it doesn't seem to be called by cymetric anywhere. This can be resolved by calling cycamore_root_metrics, but I don't see that mentioned anywhere in the docs or examples as something to do, even though cycamore arechetypes are probably used in nearly every Cyclus simulation.

Similarly, the TimeSeries tables can all be added by following the same steps manually, e.g.

timeseriesdemandcommod = root_metric(name='TimeSeriesdemandcommod')

but I think this is something we could/should do automatically during the db setup