core: add better logging for autoconnect with a new log domain

Sometimes users report problems about profiles not autoconnecting and
with the current logging it's difficult to understand where the
problem is. Since all profiles must be evaluated when a device becomes
available for autoconnection, adding by default a log message for each
of them would be probably too verbose, even for TRACE level.

Add a new 'autoconnect' log domain that is disabled by default and is
useful to debug autoconnect problems, and use it to log why profiles
can't autoconnect.
20 jobs for bg/log-autoconnect in 33 minutes and 34 seconds (queued for 3 seconds)
latest
Status Job ID Name Coverage
  Test
passed #704485
checkpatch

00:01:33

manual #704491
allowed to fail manual
t_centos:7.5.1804
passed #704492
t_centos:7.6.1810

00:25:38

passed #704498
t_debian:10

00:18:00

passed #704497
t_debian:9

00:18:12

passed #704500
t_debian:sid

00:18:12

manual #704499
allowed to fail manual
t_debian:testing
manual #704486
allowed to fail manual
t_fedora:28
passed #704487
t_fedora:29

00:33:33

passed #704488
t_fedora:30

00:27:57

passed #704489
t_fedora:31

00:29:08

manual #704490
allowed to fail manual
t_fedora:rawhide
passed #704493
t_ubuntu:16.04

00:20:43

passed #704494
t_ubuntu:18.04

00:22:00

manual #704496
allowed to fail manual
t_ubuntu:devel
manual #704495
allowed to fail manual
t_ubuntu:rolling
 
  External
failed https://desktopqe-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/job/beaker-NetworkManager-gitlab-trigger-code-upstream/1269/

passed https://desktopqe-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/job/beaker-NetworkManager-gitlab-trigger-code-upstream/844/

failed https://desktopqe-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/job/beaker-NetworkManager-gitlab-trigger-code-upstream/1269/

05:33:49

passed https://desktopqe-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/job/beaker-NetworkManager-gitlab-trigger-code-upstream/844/

06:37:08