Latest trajectory calculations reveal inconsistencies in projections and reference

[MichaelTiemannOSC]

This change reveals that the reference file may need to be updated.

857c789

Two things stand out:

1. the original reference file is happy to make projections that have no trends driving them forward (all years 2022-2050 the same)
2. When we see S1 and S2 data, the projection code should sum them if we don't also have S1S2 projections.

[MichaelTiemannOSC]

My comment was more about the reference JSON file, which is perhaps not correct in what it is claiming as a valid projection. If it looks like the projection code is now working better than before, the reference JSON file should be updated to validate the better answers. If you find a problem in the way the projection code is working (under some interpretation), then please open an issue for that! Where the projection code and reference file agree, there is no need to fix.