Add a type matching rule to the policies in question to either match or not match the types you want to exclude.
But those matching rules will still get tried for all kinds of objects that are configured for interception. And this is that additional overhead (type matching rule will have to check the type of objects being resolved) I wand to get rid of..
If I could specify a particular policy name when configuring the interception for a type, that would do the trick. Here is how it could look like:
I.e. when configuring the interceptor for Interceptable I simply tell the container that I want only policy named "ANTOHERPolicyName" to be applied to instances of Interceptable (all other policies get skipped).
Is there any way of doing so?