refactor: historical clean energy is additional, not reference

[danielolsen]

Purpose

Based on conversations with @danlivengood, we want to modify how external clean energy references are incorporated in calculations. The biggest example of this is California: they calculate a 'clean energy total' vastly in excess of our calculated numbers, since they count things like small hydro, rooftop PV, etc. that are not counted by our framework.

Previously, we had been inputting a constant value as the 'historical clean energy reference' (e.g. from California's reported total in 2016). This change instead uses a 'historical additional clean energy' value (which we will calculate once outside of this framework). This has the benefit of allowing accurate calculation of shortfall in scaled-up 2030 scenarios, where previously the over-generation was clipped at this historical reference value, even for future scenarios.

What is the code doing

In clean_capacity_scaling.py: we rename 'external_ce_historical_amount' to 'external_ce_addl_historical_amount'. This offset is added to prev_ce_generation (generated energy from 'qualifying' resources in our framework) when calculating total shortfall or overgeneration for each target area.

In test_resource_target_manager.py: we modify the test input values to reflect that the final argument to TargetManager is an offset, not a reference.

In test_strategies.py: we modify the test input values to reflect that the final argument to TargetManager is an offset, not a reference.

In test_target_manager_input.py: we modify the column naming.

Time to review

Half an hour. This change is necessary to begin our revised Western 2030 runs, which are a blocker for our USA 2030 runs.