Message retention purge jobs can incorrectly think there's no event to purge from a room

[babolivier]

An interesting bug I found while investigating #6683:

When figuring out which topological token to start a purge job at, we need to do the following:

1. Figure out a timestamp before which events will be purged
2. Select the first stream ordering after that timestamp
3. Select info about the first event after that stream ordering
4. Build a topological token from that info

In some situations (e.g. quiet rooms with a short max_lifetime), there not be an event after the stream ordering at step 3, therefore we abort the purge with the error No event found.

[babolivier]

Fixed in #6714