Update data source dtype coercion

[brynpickering]

Fixes issue where data sources contain numeric data with separators.

E.g., 25,000 would be loaded by stored in CSV as "25,000" and pandas can't natively coerce that to a numeric dtype.

Summary of changes in this pull request:

• Data that should be numeric but is stored in the array as a string due to a separator will be coerced to a numeric value according to the "locale" of the user's machine. I believe this defaults to , being the thousand separator and . the decimal point. It would probably be feasible for a user to change the locale manually ( locale.setlocale(..)) if they had data using other separators.

Reviewer checklist:

• [ ] Test(s) added to cover contribution
• [ ] Documentation updated
• [ ] Changelog updated
• [ ] Coverage maintained or improved

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (6db0657) 95.76% compared to head (9625db9) 95.77%. Report is 1 commits behind head on main.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #560 +/- ## ======================================= Coverage 95.76% 95.77% ======================================= Files 24 24 Lines 3566 3573 +7 Branches 777 713 -64 ======================================= + Hits 3415 3422 +7 Misses 86 86 Partials 65 65 ``` | [Files](https://app.codecov.io/gh/calliope-project/calliope/pull/560?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=calliope-project) | Coverage Δ | | |---|---|---| | [src/calliope/preprocess/data\_sources.py](https://app.codecov.io/gh/calliope-project/calliope/pull/560?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=calliope-project#diff-c3JjL2NhbGxpb3BlL3ByZXByb2Nlc3MvZGF0YV9zb3VyY2VzLnB5) | `100.00% <100.00%> (ø)` | |

[brynpickering]

Closing as, after some testing, locale setting turns out to not be that powerful. E.g., swiss locale can't handle 1'000.0 and german locale can't handle 1.000,0. Users will just need to sanitise their Excel data before loading it into Calliope.