create_cost_projections method causes Pandas Downcasting behavior in "replace" warning

[Tyler-lc]

Also, once you have it working properly, this happens:

[c:\***\***message-ix-models\message_ix_models\tools\costs\regional_differentiation.py:125](file:///C:/vs_code/hyway_ssp/message-ix-models/message_ix_models/tools/costs/regional_differentiation.py:125): FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version. To retain the old behavior, explicitly call `result.infer_objects(copy=False)`. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)

I think this is a relatively new warning from pandas. But it is worth looking into it I guess.

[khaeru]

A couple things about tracking issues:

• Parent/sub-issues should be used only when there is an actual relationship between two issues.

  In this case, the current issue/warning comes from the module .tools.costs.regional_differentiation. This happens to be the same module (and function) that reads the file mentioned in ValueError in get_weo_data() from PyPI distribution due to Git LFS pointers #313. However, as you describe, this warning occurs even if issue 313 is solved/not encountered. So while you may have encountered this issue in the course of (or after) investigating the other one, they seem independent.

  In other words, fixing this issue is not a necessary or sufficient condition for fixing issue 313. For that reason, it should not be a sub-issue.

• A good, explanatory title can help.

  In this case, a title that describes the issue itself (and not the parent that, per the above point, should not be the parent in any case) might be: FutureWarning from pandas in get_weo_data(). This describes what is happening and where.