fix: async client uses fixed grace period

[daniel-sanche]

Previously, when a channel refresh occurs, the async client would use channel.close() with a grace parameter to allow previous channels to keep serving old requests for a time. We were seeing flakes in our tests, showing that channel.close() isn't reliable, and can sometimes cancel ongoing requests before the grace period ends

This PR fixes this by using a fixed sleep time before calling close in the async client, like the sync client already does. This should remove the potential for cancelled requests before the grace period ends, and improve test flakiness

I also updated the system test to fully capture this problematic state, instead of encountering it in a random race condition