Investigate updating `gunicorn` worker counts

### Proposal

@cdrini points out that many of our `gunicorn` workers are sitting idle waiting for requests. As per [`gunicorn`'s guidance](https://docs.gunicorn.org/en/latest/design.html#how-many-workers), we may be using too many workers.

Let's investigate lower the number of `gunicorn` workers in our servers.

### Justification

Having too many idle workers can negatively affect site performance.

Adding more [statsD instrumentation](https://docs.gunicorn.org/en/latest/instrumentation.html) could help us determine if we should adjust the worker count, and by how much.

### Breakdown

#### Requirements Checklist

* [ ]

#### Related files

*

#### Stakeholders

*

#### Instructions for Contributors

Please [run these commands](https://github.com/internetarchive/openlibrary/wiki/Git-Cheat-Sheet#working-on-your-branch) to ensure your repository is up to date **before** [creating a new branch](https://github.com/internetarchive/openlibrary/wiki/Git-Cheat-Sheet#making-changes-and-creating-a-pull-request) to work on this issue and **each time after** pushing code to Github, because the pre-commit bot may add commits to your PRs upstream.

The issue names no files, tests, or entry points. Start by reviewing the current Gunicorn server configuration alongside Gunicorn's worker-count and statsD instrumentation guidance, then measure worker utilization and site performance. Done means a documented worker-count recommendation supported by measurements, with any needed instrumentation identified.

Written by the indexing model from the issue text.