Event processing delay can cause issue to be reopened as regression

[kupsef]

Environment

self-hosted (onpremise deployment)

Version

21.10.0

Steps to Reproduce

1/ create issue
2/ resolve issue
3/ send event for issue
4/ resolve issue right after step 3

I've done this as step 3 and 4, both scripts using the API:
./send-event.sh; sleep 0.7; close-issue.sh

sleep 0.7 is not necessary, but for me it happened consistently even with sleep 0.7, but this can ofc depend on the environment/load/etc...

Expected Result

After step 4 the issue is in resolved state.

Actual Result

Sentry marks the issue as regression.

[getsentry-release]

Routing to @getsentry/open-source for triage. ⏲️

[getsentry-release]

Routing to @getsentry/workflow for triage. ⏲️

[chadwhitacre]

Looks like @markstory believes #30372 to fall on Workflow, so syncing this as well.

[wedamija]

The proposed patch looks good to me, we just need a test to validate