Pull request latency explained: an empirical overview