Refactor GraphHandler.java and QueryRpc.java to share query logic

[HugoMFernandes]

Right now, GraphHandler.doGraph (

opentsdb/src/tsd/GraphHandler.java

Line 126 in 08acb73

private void doGraph(final TSDB tsdb, final HttpQuery query)

) implements its own query-processing logic which, among other things, fetches all series synchronously (see the TODO in

opentsdb/src/tsd/GraphHandler.java

Line 193 in 08acb73

// TODO(tsuna): Optimization: run each query in parallel.

).

QueryRpc.handleQuery has similar logic, but actually does things asynchronously (

opentsdb/src/tsd/QueryRpc.java

Line 112 in 08acb73

private void handleQuery(final TSDB tsdb, final HttpQuery query) {

).

Because of this (code path) difference, adding query-related features requires implementing the logic in both places (which has led to bugs).

I propose rewriting this block of code (i.e. the query-related logic contained in QueryRpc) to process queries/annotations/etc and return data, moving it away from handleQuery. The responsibility of QueryRpc.handleQuery would then simply be to attach a formatter to the data, while the responsibility of GraphHandler.doGraph would be to turn the data into a graph (via gnuplot).

[johann8384]

Changed to bug at the request of the @HugoMFernandes

[HugoMFernandes]

This one was fine as enhancement. My bug comment was related to the other one (#773). Anyways, it doesn't really matter.