Description
Improve our handling (or logging) for nodes that are behind and can't process Flashblocks, e.g. nodes that start from a snapshot. Currently we log:
ERROR: could not process Flashblock {missing canonical header for block XYZ. This can be ignored if the node has recently restarted, restored from a snapshot or is still syncing.}
We should detect this case (i.e. canonical chain is far behind tip) and not log errors as this is expected behavior.