fix: stop forwarding deployed tag from config:set restart

Reverts the earlier triggerRestart change. Forwarding the numeric tag forces
release_and_deploy to look up `dokku/<app>:N` (the bare local name), which
was never tagged when image-repo overrides the default. `fn-verify-app-image`
then retag-after-pulls and the underlying image ID ends up with two tags,
which prevents scheduler-docker-local's image retire path from removing the
old numeric tag (it uses unforced `docker image remove`).

plugins/config/config.go

@@ -144,12 +144,10 @@ func triggerRestart(appName string) {
 		return
 	}
 
-	imageTag, _ := common.GetRunningImageTag(appName, "")
-
 	common.LogInfo1(fmt.Sprintf("Restarting app %s", appName))
 	_, err := common.CallPlugnTrigger(common.PlugnTriggerInput{
 		Trigger:     "release-and-deploy",
-		Args:        []string{appName, imageTag, "config-redeploy"},
+		Args:        []string{appName, "", "config-redeploy"},
 		StreamStdio: true,
 	})
 	if err != nil {

tests/unit/registry_2.bats

@@ -99,22 +99,3 @@ setup_push_on_release_app() {
   assert_success
 }
 
-@test "(registry) [recover] config:set forwards deployed tag to release-and-deploy" {
-  setup_push_on_release_app
-
-  run /bin/bash -c "dokku config:set $TEST_APP RECOVERY_KEY=value"
-  echo "output: $output"
-  echo "status: $status"
-  assert_success
-  # triggerRestart forwards the deployed numeric tag (not "latest") to
-  # release-and-deploy, so the wrapper attempts recovery against ":1" rather
-  # than skipping recovery for ":latest" which is never pushed.
-  assert_output_contains "dokku/$TEST_APP:1 not found locally, pulling from registry"
-
-  # The wrapper retags the registry image back to the bare local name so
-  # subsequent get_app_image_name inspects resolve.
-  run /bin/bash -c "docker image inspect dokku/$TEST_APP:1"
-  echo "output: $output"
-  echo "status: $status"
-  assert_success
-}