AdbcStatementExecuteUpdate/AdbcStatementExecuteQuery & rows_affected

[hannes]

AdbcStatementExecuteUpdate can set rows_affected, which is a bit ancient given that most SQL systems support the RETURNING clause now, which makes the separation between ExecuteQuery and ExecuteUpdate redundant. I propose to have only one method to run a query, possibly along with allowing the out to be nullptr in AdbcStatementExecuteQuery. rows_affected should be removed.

[hannes]

See also https://www.sqlite.org/lang_returning.html

[lidavidm]

Whoops, I didn't see this this morning. Folding Query/Update together sounds reasonable, leaving Partition separate. (Sorry for the churn.)

The explicit rows_affected, parameter, though, helps when trying to adapt to things like Python's DBAPI and Go's database/sql (as @zeroshade pointed out). So I don't think it hurts to have it, even if support might be a mixed bag.

[lidavidm]

See #95