Seed multiple content items per id closeness

[kdeme]

Once we have https://github.com/status-im/nimbus-eth1/issues/1080 we can start with experimenting on how to efficiently group (and potentially regroup on NH gossip) content items per id closeness.

A good approach could be to start experimenting with this first on the initial seeding/injecting of data into the network, and only later tackle this in the NH gossip. This items is about that, improving the current 1 by 1 content item seeding.