Extend BPEiC and PERAC multipliers to time series

[jrissman]

To allow regions that have their own detailed process emissions data (for BPEiC) but no abatement data (for PERAC) to use their BPEiC data instead of the data in the built-in international dataset, we need to extend the existing multiplier system. Multipliers must be a time series and should be disaggregated by gas (in addition to their current disaggregation by sector). Multipliers in BPEiC should then be set to cause the region's data from the built-in international dataset to be scaled to exactly match the detailed region-specific process emissions data, and those same multipliers should be used in PERAC.

[jrissman]

Commits 5bc52c0 and 42af6fc implement this for the U.S. model.

Note that after discussion with Sonia, we're going to extend to time-series but not provide separate multipliers by gas, as it's not worth the complexity, and the purpose of the non-country multipliers is to adapt the model to a non-country region, not to allow for arbitrary manipulation of the EPA dataset.