chore: cache error tracking query longer

[daibhin]

Problem

Error tracking queries are not being cached for long enough

Changes

Without the query_date_range the cache_target_age interval gets set to minute, which results in a 15 second cache. Bumping the interval to hour will cache for 15 minutes which feels more reasonable for the error tracking query

[neilkakkar]

something to eventually think about: One workflow I love with sentry is:

1. I go into a user account to try and repro a bug. This throws an error
2. I immediately go into sentry to find the issues in the last few minutes to correlate and find the right one (maybe some searches as well, but it can sometimes be hard to figure out what to search for, like in async cohort recalculation celery tasks)
3. Sentry shows me the exact issue and I can go from there

With caching for 15 minutes at a time, this workflow becomes much more painful...