Add `DQR_REQUEST` label to trigger the DQR automatically for significant early warning triggers not associated with full bandwidth ones

At the moment, the DQR_REQUEST label is not added for a significant early warning trigger. This is fine for the case when a full-bandwidth trigger is going to complement that early warning trigger, because its processing will lead to DQR_REQUEST being applied in due time. On the other hand, if the early warning trigger remains orphan from a full-bandwith trigger, the instructions are to retract the associated public alert. To do that with more confidence, it would be good to have DQR results to look at -- and hence to have the DQR_REQUEST` label be applied automatically at some point in time.

=> Would it be possible to implement the following algorithm?

If a significant early warning trigger has no associated full-bandwidth trigger after N minutes (with N being small but larger than the latency of low-latency pipelines to upload their triggers; N=5 may work), add the DQR_REQUEST label to its S-event, in order to trigger the running of the DQR at the different sites.

For the record, tagging @keita.kawabe , @jenne-driggers , @tsawada and @francesco.direnzo .