Do area-preserving interpolation

[chrisroadmap]

For the country-level and global data in the GFED emissions we use linear interpolation to shift from the 0.25° emissions grid to the 0.5° ISO grid.

https://github.com/iiasa/emissions_harmonization_historical/blob/1994d4b1be44b4bd8f07ef45e3dc08da6513ce4d/notebooks/0102_GFED4-prepare.py#L156C1-L159C92

The regridding should be area weighted rather than linear. I verify that I get slightly different global total emissions when I have done this previously in processing CEDS emissions, and openclimate data also agrees with my method.

Proposal is to use iris and the area weighted regridding with that package.

Will be included in the #21 PR.