Thomas Bachem's first patch notes that the documentation for hook.<event>.enabled, added in dcfb5af67e, says that with the setting false "no hooks fire for that event". In practice only hooks defined in configuration are skipped. With hook.pre-commit.enabled=false, git commit still runs .git/hooks/pre-commit, and git hook list shows it without the "event-disabled" mark. His patch moves the event_disabled flag out of u.configured so it applies to both kinds of hooks.
The second patch concerns hook.<name>.enabled and hook.<name>.parallel. These are booleans, and git-config documents a valueless boolean as true. But hook_config_lookup_all() rejects any hook.<name>.<key> with no value, so "git -c hook.lint.parallel hook run pre-commit" fails with "missing value". The patch keeps rejecting missing values for other keys and passes these two to git_parse_maybe_bool().
Junio C Hamano said he suspects the current behavior of the first issue is deliberate and the documentation was written too loosely, as if configuration-defined hooks were the only kind. He said he wanted to see what others want.
Adrian Ratiu, the original author, agreed with that reading. He said he did not intend to disable hookdir hooks via config, since his reasoning was that configs affect only config-defined hooks and traditional hooks behave as before. He said the documentation is too loose, but he has no strong opinion and is fine accepting the patch if others think it a good idea.
So the choice between changing the behavior and fixing the documentation is open. The excerpts show no discussion yet of the second patch.