Allow all warning types in manually created INKY rules

Currently, when adding entries manually to the allow list, there's only a limited selection of options to choose from, leaving several unable to be selected for preparatory rules. For example, "Brand Impersonation" rules are unable to be selected when creating a rule. Instead, an email has to be actively blocked because of impersonation, and only then can we make a rule. This means there has to be some amount of issue before we can remedy, and it can't be fixed beforehand even if we know about the expected problem.

Please allow us to create rules for all available kinds of analysis results.