#8 Batching and throttling for MetDriver

[Egon Willighagen]

> (Things may evolve a bit... suggestions most welcome...)
And some throttling mechanism, which avoids 1500 queries/sec
on e.g. some xcms result set ;-) For the results I suggest
to cache at least in memory, so a repeated query will not
go to the web service again.

And also to allow a batch mode, which e.g. PubChem allows.
Maybe call it IMetaboliteDatabaseCachingDriver ?

public interface IMetaboliteDatabaseDriver {
// Database name
public String getName();

// Metabolite queries, with batch processing
public void begin();
public void queryMetabolites(float mass, float range);
public void commit();

// Metabolite queries