Schema cleanup

[zaneselvans]

Overview

Some miscellaneous column mapping cleanup, mostly for the entity-specific columns which get turned into operating_date values. Things I came across while looking into a better fix for #423.

• Map the operating_year and operating_month columns in the EIA860M retired generators consistently -- these were getting silently lost because the other EIA860 generator operating date columns are name generator_operating_date.
• Ensure all *_(operating|retired)_(year|month) columns have specific names in column mapping metadata CSVs, and remove the manual renames that were happening elsewhere to ensure that there aren't a bunch of different operating_date and retirement_date columns floating around in the transforms which actually apply to different entities. This change affected the operating month/year/date columns for cooling_system, emission_control, fgd, and fgp.
• Put the convert_to_date() and month_year_to_date() helper functions next to each other in their module... they really need to be combined!
• Rename obstacle_num_faa to obstacle_id_faa and update field description to provide more context on what those IDs refer to.
• Rename core_eia860__yearly_generators_wind to core_eia860__scd_generators_wind, as it's not a data table, but a generator attribute table.
• Add a new alembic migration to deal with the wind table and FAA column renames.

Testing

• I ran the full ETL locally and didn't get any failures due to bad / missing column or table names.
• I ran make pytest-validate on the fully populated database and it passed.
• I ran make pytest-integration-full on the fully populated database and it passed.

To-do list

Preview Give feedback

1. Update the release notes: reference the PR and related issues.
2. Ensure docs build, unit & integration tests, and test coverage pass locally with make pytest-coverage (otherwise the merge queue may reject your PR)
3. Review the PR yourself and call out any questions or issues you have

[cmgosnell]

📜 🧼 👍🏻