Hot fix of some issues

[davidrzhdu]

The issues get resolved in this PR are:

1. Add .load() to every call of EconVars in dscim.preprocessing.input_damages.py. This is for resolving the chunk issue.
2. Convert output of dscim.preprocessing.input_damages.compute_ag_damages() to float32 to make Ag consistent with other sectors.
3. Convert list passed to pandas.DataFrame.groupby() in dscim.utils.rff.py to string when grouping pandas.DataFrame by a single column. Passing list containing a single string to pandas.DataFrame.groupby() works well locally, but creates some weirdness on GitHub.

[codecov]

Codecov Report

Merging #82 (7bf37fa) into dscim-v0.4.0 (f5d88b2) will decrease coverage by 0.61%.
The diff coverage is 0.00%.

@@               Coverage Diff                @@
##           dscim-v0.4.0      #82      +/-   ##
================================================
- Coverage         40.53%   39.92%   -0.61%     
================================================
  Files                18       17       -1     
  Lines              1838     1866      +28     
================================================
  Hits                745      745              
- Misses             1093     1121      +28     

Impacted Files	Coverage Δ
src/dscim/preprocessing/input_damages.py	0.00% <0.00%> (ø)
src/dscim/utils/rff.py	0.00% <0.00%> (ø)

... and 2 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[kemccusker]

Thanks for these fixes!