This patch looks good to me. I have two very minor nits: The inflation of the sample size by 10% is arbitrary but it
doesn'tseem unreasonable or concerning. It just makes me curious if there are any known cases that motivated adding
thislogic. Secondly, if the earliest non-deprecated version of PostgreSQL supports sampling, then you could optionally
removethe logic that tests for that. The affected lines should be unreachable.
