Please report some of these cases to the list. We're much more likely to fix things that we hear are practical problems than the ones that we know theoretically

Please report some of these cases to the list. We're much more likely to fix things that we hear are practical problems than the ones that we know theoretically exist, but are much more likely waste of plan time for everyone, not to speak of development and maintenance overhead. I find "disable sequential scans entirely proves to me that Postgres doesn't care about performance for interactive use." in combination with not reporting the issue a bit contradictory. We care about stuff we get diagnosable reports for.