In SERVER-112273 I fixed a bug where a user could use the killOp() command to kill internal repl thread (we've seen it happen with OplogWriter), and along this fix I added a new jstest for it in killOp_against_repl_threads.js. In this test I originally wanted to try to kill the OplogWriter and OplogApplier operations. But the OplogApplier operation can be created and deleted (and then re-created) within 1sec which made the test flaky. So I tried to re-use the rsSyncApplyStop FP to make the OplogApplier thread always stop on this FP when enabled. But this caused BF-41309 once merged. To fix it I reverted the change to always stop on the rsSyncApplyStop FP even when featureFlagReduceMajorityWriteLatency is enabled, and removed the test for trying to kill the OplogApplier operation in killOp_against_repl_threads.js. This was done as part of SERVER-118535.

We should re-add the test to check that the OplogApplier operation cannot get interrupted by killOp() in killOp_against_repl_threads.js. In order to do that we should introduce a new FP in OplogApplier.