Comment # 13 on bug 1001790 from Franck Bui

Weird how this was handled and sadly I wasnt notified (as author of the
patch/regression).

So the concept of "IgnoreInibit works for high-level locks only" was totally
new to me and probably for a lot of people since the documention never
mentionned it (and explains why it was accepted upstream without any issue).

It's also not clear to me why we still have and use
"LidSwitchIgnoreInhibited=yes" but anyways I'm going to backport the revert of
the patch and the change in the documentation to match the restored behavior.