confbridge: Fix a bug which made conferences not record with AMI/CLI commands

When confbridge was changed to handle conference status with a state machine in
r374658. The function responsible for starting recording for a conference was
refactored with the function actually responsible for launching the recording
thread being split into a function with another name. The old function name was
still used for manually started recordings through AMI or CLI. This patch fixes
that by switching which function is used to start recording the conference.

(closes issue ASTERISK-20601)
Reported by: Vilius
Patches:
    confbridge_mixmonitor.diff uploaded by Jonathan Rose (license 6182)
........

Merged revisions 375470 from http://svn.asterisk.org/svn/asterisk/branches/10
........

Merged revisions 375471 from http://svn.asterisk.org/svn/asterisk/branches/11


git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@375472 65c4cc65-6c06-0410-ace0-fbb531ad65f3

apps/app_confbridge.c

@@ -2273,7 +2273,7 @@ static char *handle_cli_confbridge_start_record(struct ast_cli_entry *e, int cmd
 		ast_copy_string(bridge->b_profile.rec_file, rec_file, sizeof(bridge->b_profile.rec_file));
 	}
 
-	if (conf_start_record(bridge)) {
+	if (start_conf_record_thread(bridge)) {
 		ast_cli(a->fd, "Could not start recording due to internal error.\n");
 		ao2_unlock(bridge);
 		ao2_ref(bridge, -1);
@@ -2611,7 +2611,7 @@ static int action_confbridgestartrecord(struct mansession *s, const struct messa
 		ast_copy_string(bridge->b_profile.rec_file, recordfile, sizeof(bridge->b_profile.rec_file));
 	}
 
-	if (conf_start_record(bridge)) {
+	if (start_conf_record_thread(bridge)) {
 		astman_send_error(s, m, "Internal error starting conference recording.");
 		ao2_unlock(bridge);
 		ao2_ref(bridge, -1);