Maintain buckets according to specs

[herr-seppia]

Describe what you want implemented
The bucket periodic maintenance need to follow the specs
According to white paper:

• Each node periodically refreshes every bucket it has not seen some activity from in the last hour: for each such bucket, it picks a random ID with appropriate distance and performs a look up to populate its buckets with fresh routing information.
• The lookup procedure allows a node to retrieve a set of k nodes closest to a specific ID in the address space.
• The procedure of finding the k closest nodes is carried out by iteratively narrowing down the search space and issuing FIND_NODE messages to nodes which are closer to the ID.

Describe "Why" this is needed
Actually the nodes lookup maintenance is performed against a random known peer's ID inside the bucket instead of a random ID

Describe alternatives you've considered
Keep going with the actual implementation

Additional context
To handle properly this issue, the issue #44 should be resolved first