Add an LTI 1.3 debug log in the case that the JWT fails to decode on a launch request.

In this case the LTIAuthenError stash value is set, but that error is
never shown anywhere because the course ID was not determined and the
authen verify method is never called.  So this information goes to the
abyss and the debug log abrubtly terminates (see
https://forums.openwebwork.org/mod/forum/discuss.php?d=8749 for what
this debug log looks like).  So this pull request always debug logs it
in this case.  There is not enough information to even determine if LTI
debugging is enabled for the course (which the request failed to
determine), so it has to be the more general debug logging facility.
This may help in resolving the issues that are occuring with the cases
such as those in the above mentioned forum post and
https://forums.openwebwork.org/mod/forum/discuss.php?d=8738#p22228.

Author: drgrice1

lib/WeBWorK/ContentGenerator/LTIAdvantage.pm

@@ -161,6 +161,8 @@ sub launch ($c) {
 					$c->stash->{lti_jwt_claims}{'https://purl.imsglobal.org/spec/lti/claim/context'}{id}
 				]
 			];
+		} elsif ($c->stash->{LTIAuthenError}) {
+			debug($c->stash->{LTIAuthenError});
 		}
 		return $c->render(
 			'ContentGenerator/LTI/content_item_selection_error',