In a couple of places ("cluster show" and "service status") we printed
connection status errors as a "WARNING" to stdout, followed by a log
HINT, but the latter was ineffective unless the configured log
level happened to match the level of the most recently emitted log
line (which will most likely be DEBUG).
Convert the printed WARNING lines to an actual log WARNING, to make the
behaviour of this output behave as expected.
"repmgr daemon" can be interpreted to mean the commands affect the local
daemon process only. Rename the commands which affect the entire cluster
to "repmgr service ...".
The "repmgr daemon ..." form of the affected commands is retained for backwards
compatibility.
