Revamp `pudl-usage-metrics` repo

[bendnorman]

We have a goal this season to start monthly user metrics calls with the team. The dashboard should include data from

• s3 bucket logs
• Cloud Run and Fly.io logs
• Superset logs

Data cleaning tasks

Beta Give feedback

1. Port datasette logs to a cloud bucket.
2. Figure out how to save logs and users from superset
3. Rip out Intake cleaning code
4. Create an ETL for s3 logs processing
5. Add google analytics metrics
6. Add superset metrics: what metrics do we want? Number of new sign ups? Number of active users?

We should also probably revamp out pudl-usage metrics repo because it's quite stale at this point!

Infrastructure tasks

Beta Give feedback

1. Update packages #119
2. The repo was created with an older version of the cheshire template. Update the packaging methodology and pre-commit hooks. Replace setup.py with pyproject.toml and conda for mamba. #119
3. Update Python version to 3.12 #119
4. The repo uses sqlite for local development. Duckdb might be a better option at this point.
5. Maybe use Bigquery instead of cloudsql because it's cheaper and faster
6. The repo was written back when ops were the primary abstraction in dagster. Rewrite the code to use @assets. We could also rework the repo structure to reflect the change of abstractions.
7. Re-evaluate use of dagster partitions. Our usage metrics repo is a good fit for the partitions feature because we just want to execute a single partition of data every day. I remember running backfill was a time consuming process because dagster launched a process for every single partition (day). Does it still make sense to use partitions? Should we adjust the frequency?

[e-belfer]

To reduce duplication between this issue, I've migrated the content from this issue into #128 and transformed it into a complete epic. Closing this issue to reduce duplication.