Notification Grace Period may be ignored in certain conditions

Event notification grace period appears to be getting ignored under certain conditions.

I observed this and reproduced (see steps to reproduce).


## Expected Behavior

Notification grace period is not ignored

## Current Behavior

Notification grace period is ignored.

## Possible Solution

## Steps to Reproduce (for bugs)

1. Configure message to send to a Graylog Forwarder
2. Create a valid notification that can send/execute
3. Create/configure an event with the following
*
* configure the notification created in the previous step, configure a grace period that you can test with
4. With the messages being sent to the forwarder, stop Graylog service (e.g. `systemctl stop graylog-server`)
5. wait an amount of time that is a multiple of the grace period, for example, if your grace period is 5m, wait 10m or 15m
6. re-start Graylog service
7. Observe that forwarder reconnects to graylog and all of its messages are dumped into `graylog-server`'s journal.
8. As messages are processed, the event criteria will be met, and several notifications will be sent all at the same exact time, ignoring the grace period

## Context

I originally encountered this as my graylog node was mistakenly offline for a couple of days and received a large amount of notifications all at once when [re]starting graylog.

## Your Environment

* Graylog Version: 6.1
* Java Version: bundleed
* OpenSearch Version: 2.15.0
* MongoDB Version: 7.0.17
* Operating System: Ubuntu Server 22.04 LTS
* Browser version: Google Chrome Version 134.0.6998.118 (Official Build) (arm64)

Please let me know if there are any questions.