Install Councilmatic in pupa settings file

[hancush]

Overview

A long time ago, in a galaxy far away, we used one scraper repository for many jurisdictional scrapers. These scrapers shared a pupa settings file that needed to account for some jurisdictions scraping into a plain OCD database, while others scraped into a Councilmatic database. Thus, we introduced a SHARED_DB environment variable that, if true, would include councilmatic_core in installed apps.

The script we use to run scrapes sets the SHARED_DB variable, but it was not set for the pupa clean DAG, so the councilmatic_core app was not installed. This resulted in the deletion of top-level entities like organizations and people behaving badly, i.e., not cascading as configured in the Councilmatic app.

This PR does away with the conditional behavior in the settings file. The Metro scrapers are always running against a shared DB.

Connects datamade/django-councilmatic#293

Notes

Once this is merged and the main tag builds, we should be able to run the cleanup DAG on the staging dashboard without error.

Testing Instructions

Ran locally against the staging database to confirm this works!

docker run -it -e DATABASE_URL=${STAGING_DATABASE_URI} scrapers-lametro /bin/bash
# inside container
pupa clean