improve performance reading big resultset

Actual implementation check column metadata for each decoding value :
getInt() will check column medatadata to check column datatype, signed/unsigned value, ... for each decoding.

When having lots of rows, that represent a significant amount of time.
A better solution would be to set a decoding method reference depending on datatype, permitting to decode value.
This need benchmark, because method reference use might prevent better performant