Add optional bucket parameter to list keys, log siblings and resolve siblings

[dmitrizagidulin]

When the data set gets large enough, the sibling-related operations (log siblings and resolve siblings) take too long and put too much load on the cluster (since they require iterating through each riak object, not just each key).

Add an optional parameter to the count_all_keys(), resolve_all_siblings(), and log_all_keys() functions to restrict the operation to a particular bucket (or a bucket type/bucket pair). Take a look at how the SleepPeriod optional parameter works in line 52-53

[rclar2s]

This is what they posted in the ticket 11116 in terms of new requirements

1. Applying the tool on particular set of data
   for example: Bucket level
2. Tool accepting list of keys, so that the resolution happens only in that list of keys provided.

[nerophon]

Did the first part, to work on specified buckets.
Did not yet do part two, to work on a list of keys.