Regression version comparison incorrectly identified issue as regressed, when version was older than resolved-in version

[jboulter11]

Environment

SaaS (https://sentry.io/)

Steps to Reproduce

https://dropbox.sentry.io/issues/4722460418/activity/?project=4505546600087552
https://dropbox.sentry.io/issues/4704784157/activity/?project=4505546600087552
https://dropbox.sentry.io/issues/4722450385/?project=4505546600087552

All three of these issues opened back up with regression versions which were older (smaller version number) than they were resolved in. This creates noise and leads to lack of trust in Sentry's regression detection.

This is not the first time we've had this issue. #74448

This is similar to, but not exactly the same as this issue: #89191
The three Sentry issues I linked above do not say compared via release date but I figured I'd mention this other open ticket just in case.

Expected Result

Issues would stay closed because no version newer than the resolved version has reported events to the issue group.

Actual Result

The issues were re-opened.

Product Area

Issues

Link

No response

DSN

No response

Version

No response

[linear]

ENG-5782

[getsantry]

Auto-routing to @getsentry/product-owners-issues for triage ⏲️

[jboulter11]

Another example of this https://dropbox.sentry.io/issues/4733508787/activity/?project=4505546602381312

[scttcper]

I think the issue is that the jira resolution function wasn't setting the resolution id on the activity entry. This caused the regression activity to never get updated with all the information we need to determine what was compared during the regression. We should be able to display what releases were compared and if we used semver or datetime to compare them. I've put up an update to fix that on future resolutions via jira and another to start displaying the integration that was used to resolve an issue.