Re-open Semantic Logger appenders in Cluster mode

[mwtrew]

This missing config meant Semantic Logger was not working in prod, where we use Puma in Cluster mode.

See https://logger.rocketjob.io/forking.html#puma

Status

• Related to https://github.com/RaspberryPiFoundation/digital-editor-issues/issues/1455

What's changed?

• Adds config so that Semantic Logger works when we're using Puma with multiple threads, like we do in production. See the docs for more info.

Steps to perform after deploying to production

None

[Copilot]

Pull request overview

Ensures Semantic Logger appenders are correctly re-opened after Puma forks worker processes in clustered mode so logging works in production.

Changes:

• Adds a Puma worker lifecycle hook intended to call SemanticLogger.reopen after forking.
• Minor whitespace changes at the end of config/puma.rb.

Comments suppressed due to low confidence (1)

config/puma.rb:58

• There are extra blank lines (and trailing whitespace) at the end of the file; this will typically trip whitespace/style checks and adds noise to diffs.

# Allow puma to be restarted by `bin/rails restart` command.
plugin :tmp_restart

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Test coverage

91.72% line coverage reported by SimpleCov.
Run: https://github.com/RaspberryPiFoundation/editor-api/actions/runs/27760616674