Persist in-memory replication stats to disk (#15594)

to avoid relying on scanner-calculated replication metrics.
This will improve the accuracy of the replication stats reported.

This PR also adds on to #15556 by handing replication
traffic that could not be queued by available workers to the 
MRF queue so that entries in `PENDING` status are healed faster.

cmd/peer-rest-client.go

@@ -263,7 +263,7 @@ func (client *peerRESTClient) GetAllBucketStats() (BucketStatsMap, error) {
 	values := make(url.Values)
 	respBody, err := client.call(peerRESTMethodGetAllBucketStats, values, nil, -1)
 	if err != nil {
-		return nil, err
+		return BucketStatsMap{}, err
 	}
 
 	bsMap := BucketStatsMap{}