The Slow SessionWorkflow log in session_workflow.cpp currently logs based on what it deems "active time". This is effectively all time that is not waiting for a message to be received.

It turns out that a lot of this active time can be time that the Slow Query log does not count, causing the SessionWorkflow log to happen way too often, introducing a performance regression (BF-26754).

Given that our criteria isn't as nuanced as it should be, and this layer of the system does not have access to the data required to make such a nuanced decision, we should revert back to the intention of this bug and simply log when `sendResponseMillis > slowMS`.

The log message itself should change and the test for the Slow SessionWorkflow log should be updated (slow_session_workflow_log.js).