Add census data

[fvankrieken]

Two things.

• primary purpose - add yml templates and python scripts for dcp-provided census data for population and housing. There are two source files - one by blocks/borough/city, and one by ntas/community districts/council districts/borough/city.
• some maintenance - a while back I edited a bunch of the scripts and made it so that path was specified in yml, not just hard-coded in a script (in the case of script-sourced dataset that are still pulling from just one endpoint). I missed bpl_libraries back then so added it now (and in the one test yml file that uses it)

[fvankrieken]

@alexrichey this is pretty limited in scope but is an example of adding new data sources via data library - two new yaml files that specify that the dataset "source" is from the added scripts

[fvankrieken]

I think I get the gist of the changes. Looks good to me, though maybe we want consider a linter/formatter to flag unused imports and enforce style?

We use black often for linting/formatting but could definitely stand to incorporate it with a tad more rigor/norms (commit hooks or something along those lines).

I don't think it complains about unused imports (or removes them)

[damonmcc]

(started typing this before I saw Finn's reply lol)

@alexrichey agree on the value of linters/formatters. we haven't taken a shot at implementing them as a CI check on all mono repo PRs, but would love to here sometime soon

black has been our go to for python formatting (it may flag unused imports though) and here's where it was being used by some datasets that are now in the mono repo

I had some fun exploring the use of trunk to lint many languages at once, but the overhead of the VS Code extension has me turning away from it