add data update tracking using bloom filter (#9208)

By monitoring PUT/DELETE and heal operations it is possible
to track changed paths and keep a bloom filter for this data. 

This can help prioritize paths to scan. The bloom filter can identify
paths that have not changed, and the few collisions will only result
in a marginal extra workload. This can be implemented on either a
bucket+(1 prefix level) with reasonable performance.

The bloom filter is set to have a false positive rate at 1% at 1M 
entries. A bloom table of this size is about ~2500 bytes when serialized.

To not force a full scan of all paths that have changed cycle bloom
filters would need to be kept, so we guarantee that dirty paths have
been scanned within cycle runs. Until cycle bloom filters have been
collected all paths are considered dirty.

cmd/peer-rest-client.go

@@ -524,6 +524,25 @@ func (client *peerRESTClient) RemoveBucketObjectLockConfig(bucket string) error
 	return nil
 }
 
+// cycleServerBloomFilter will cycle the bloom filter to start recording to index y if not already.
+// The response will contain a bloom filter starting at index x up to, but not including index y.
+// If y is 0, the response will not update y, but return the currently recorded information
+// from the current x to y-1.
+func (client *peerRESTClient) cycleServerBloomFilter(ctx context.Context, req bloomFilterRequest) (*bloomFilterResponse, error) {
+	var reader bytes.Buffer
+	err := gob.NewEncoder(&reader).Encode(req)
+	if err != nil {
+		return nil, err
+	}
+	respBody, err := client.call(peerRESTMethodCycleBloom, nil, &reader, -1)
+	if err != nil {
+		return nil, err
+	}
+	var resp bloomFilterResponse
+	defer http.DrainBody(respBody)
+	return &resp, gob.NewDecoder(respBody).Decode(&resp)
+}
+
 // SetBucketPolicy - Set bucket policy on the peer node.
 func (client *peerRESTClient) SetBucketPolicy(bucket string, bucketPolicy *policy.Policy) error {
 	values := make(url.Values)