Summary:
Original commit: d07daec2c30774f68abf0841f8ac11b3f239e39b / D39552
Switch pg_cron to normal using normal `SIGTERM` and `SIGQUIT` handlers.

In an attempt to fix stuck pg_cron workers the handler of `SIGTERM` was changed to `quickdie` 4b4c2016409901982174e9240084dddf85a94e87/D37591. But in `quickdie` mode postgres does not invoke `ReportBackgroundWorkerExit` which signals the parent process when its child backend exits. This causes pg_cron launcher to get stuck in `WaitForBackgroundWorkerShutdown`.

With 111b65d1f1a0eee4d33451596fd336c8eb5c5577/D39207 the issues with pg shutdown have been addressed and it is safe to now switch to the normal SIGTERM behavior which makes the background workers `die` in a clean manner.

#24706 tracks the issue where in yb `ReportBackgroundWorkerExit` is not called when the backend has a non graceful exit.

Fixes #24658
Jira: DB-13724

Test Plan: PgCronTest, DeactivateRunningJob

Reviewers: telgersma

Reviewed By: telgersma

Subscribers: yql, ybase

Differential Revision: https://phorge.dev.yugabyte.com/D39617