Handle query timeout errors

[AndreaNassisi]

Today the utility crashes when a query like the one below goes in timeout.
MATCH (from)-[r:]->(to) RETURN DISTINCT labels(from) as fromLabel, labels(to) as toLabel

The error message can offer the solution to change the neptune_query_timeout (cluster-level parameter).(https://docs.aws.amazon.com/neptune/latest/userguide/parameters.html#parameters-db-cluster-parameters-neptune_query_timeout)

Alternatively, the utility can fall back to small queries. Sampling each node label combination for the <edge_type>.
Is slower, but will complete with no user intervention.

[andreachild]

It may be possible to use the detailed option of the graph summary endpoint to obtain this info (or at least help to optimize subsequent queries) instead of executing this particular query causing the timeout:

https://docs.aws.amazon.com/neptune-analytics/latest/userguide/query-APIs-graph-summary.html