Query Cache | min_query_duration

As a user, I want to optimize the memory usage of the maxscale query cache to the most useful long running query results and limit consistency risk of cheap/small queries by being able to define a minimum duration in ms a query needs to run for its result to be stored in the query cache.

Today, with TTL caching, fast queries share in the consistency risk that the data might have changed with little to no performance benefit and it takes up space in the cache.

With this feature, the user can define where this breakpoint makes sense for their workload. including whether the cache is local allowing them to use a lower min_query_duration vs if the cache is behind a network hop, requiring a higher min_query_duration.