Memory efficient results.csv creation

[rajeee]

Potentially fixes #253.

Pull Request Description

Instead of loading all the results_jobx.json.gz files at once to memory, use dask to load only the results_jobs for an upgrade at a time. This should prevent out of memory error when dealing with large run with many upgrades.

Checklist

Not all may apply

• Code changes (must work)
• Tests exercising your feature/bug fix (check coverage report on CircleCI build -> Artifacts)
• All other unit tests passing
• Update validation for project config yaml file changes
• Update existing documentation
• Run a small batch run to make sure it all works (local is fine, unless an Eagle specific feature)
• Add to the changelog_dev.rst file and propose migration text in the pull request

[rajeee]

Works in a small batch. Need to verify for a super large batch that previously would have failed.

[nmerket]

Looks good from what I can see. I think the dask dataframe is a good choice here. Let me know how it goes with a larger dataset.

[rajeee]

Works in a large run with 350K buildings, 16 upgrades. With n_worker=10 for postprocessing took ~10 hours.

[nmerket]

Works in a large run with 350K buildings, 16 upgrades. With n_worker=10 for postprocessing took ~10 hours.

That's a long time, but it's good that it worked. I suppose that's what matters.