Account for code/BPS policies

Create and process new input data (./scout/supporting_data/sub_fed/codes.csv, ./scout/supporting_data/sub_fed/codes_lag.csv, and ./scout/supporting_data/sub_fed/bps.csv) to represent the effects of state and/or local new building energy codes and building performance standards (BPS) for certain market segments, beginning in certain years.

A new function, import_state_data, was written to read in and finalize the inputs for codes/BPS and state-level appliance regulations (see commit 9adac31) for further processing all in one place.

For codes, the file  provides information about the relative energy use improvements that would result from updating a given state/sector’s new building code to the latest available version. These data are drawn from the DOE codes program site, see https://docs.google.com/spreadsheets/d/1lEHsz7_66qNgxrbMV7UvAEOTjmelQpjfRwVEutK5Mzk/edit?gid=4778459#gid=4778459 and the sheet “codes_state_slopes” for more details on the derivation.

For both codes and BPS, the two key levers for reductions are a) onsite emissions reductions, and b) additional energy use savings from updating to the latest baseline code or adopting a stretch code that goes beyond the latest baseline code. The former case is operationalized as conversions of fossil-fired equipment to more efficient electric equipment, and the relative performance impact (in terms of energy/carbon/energy costs) from switching between fossil and electric units is based on data for available minimum electric efficiency measures in the analysis (e.g., the lowest-level electric switching performance available). The latter case then assesses what additional energy use reductions would be required to reach the code/BPS threshold, after having already run through all normal measure competition and assessed the energy impacts of code-/BPS-driven onsite emissions reductions.

Code/BPS effects are reflected through a new Codes_BPS_Measure object class, where one measure each represents codes and BPS and is appended to the existing measure set in the results with formatting that matches that of existing non-codes/BPS measures. At a basic level, all of the new functions are removing data from the existing non-codes/BPS measures and adding data to the codes/BPS measures, such that the specific effects of the codes/BPS measures are isolated in the results and the full measure set sums back to the correct aggregated baseline/efficient case outcomes.

Addresses issue #412.