En 7369 stop node after offline sync

[iulianpascalau]

• added the gracefully close channel in storage header resolver so the import process can safely shutdown the node when completed

Manual testing procedures:

• provide a db directory containing data in continuous epoch.
• move this directory into a new directory, might be import-db, place the directory near the config directory.
• the process can be run either with a partially build db directory or without a DB.
• change the DestinationShardAsObserver option in prefs.toml to match the containing DB directories and desired shard.
• use the same config files that have generated the initial import-db data, otherwise the node can not advance
• run the node with ./node -use-log-view -log-level *:INFO -log-save -import-db ./import-db.
• the process should iterate on each and every block in advancing in every existing epoch
• the process stops when it hits an epoch that it has a missing data (this will usually occur after traversing the whole import-db directory)

Note: please check that the node has at least 3 peer connections, otherwise the sync and consensus mechanism are paused.

[SebastianMarian]

strEpochNotFound += fmt.Sprintf(" or %d", crtEpoch-1)
or maybe if we want to print them in ascending order:
strEpochNotFound = fmt.Sprintf("%d or %d", crtEpoch-1, crtEpoch)

[SebastianMarian]

It is not crtEpoch and crtEpoch+1 ?

[iulianpascalau]

right, will change the print

[LucianMincu]

System tests passed.