Fix: Convert has_fetcher and get_fetcher_count to async

[HosseinKaviani-H]

Monarch actors require all endpoints to be either sync or async, not mixed. This change makes has_fetcher and get_fetcher_count async to match other endpoints in GlobalLoggingActor, preventing actor deadlock errors.

[felipemello1]

thanks for the PR! could you please check all the places where those are being called and see if we need to add an await?

[HosseinKaviani-H]

thanks for the PR! could you please check all the places where those are being called and see if we need to add an await?
Both methods are safe:

has_fetcher is a single call which already uses await: Location: src/forge/observability/metric_actors.py:106.
get_fetcher_count: not used anywhere in the codebase

[felipemello1]

thanks for the PR. Will merge when tests pass