Handling of pruned channels that come back to life is very inefficient

[pm47]

When we get a valid channel_update for a pruned channel, we add a QueryShortChannelIds containing a single shortChannelId to our batch. If a node has been shut down for a long time, this will result in a lot of queries and be very inefficient. Instead, we should just add the shortChannelId to an existing QueryShortChannelIds

https://github.com/ACINQ/eclair/blob/f7a79d10b4ddb6f91a4c66983c7a1620cc572e72/eclair-core/src/main/scala/fr/acinq/eclair/router/Validation.scala#L379-L389