If only the transformed can know, how could a mystic ever be shown to be wrong?

Since I campaign against superstition, let me say first that I'm not asking this to sneer. I read Sufi poetry with real pleasure, and I've learned from discussions here that the Sufi claim is subtler than "feelings beat reason". The claim, as I understand it, is that some knowledge is available only to a person who has been transformed by a discipline.

My problem is verification. If the knowledge is open only to the transformed, then nobody outside can check it. And when two practitioners report incompatible disclosures, how is the disagreement settled? The tradition's own answer, as I understand it, is that a disclosure must agree with scripture and with the reports of earlier masters. But that sends the question back to the authority of revelation, which is a different argument.

The tradition has its own live case. The doctrine later labelled wahdat al-wujud, the "unity of being" associated with Ibn Arabi (a phrase he apparently never used himself), was read by critics as erasing the distinction between Creator and creation. Defenders say the critics misunderstood what "unity" means. Both sides include serious, practising people. So even inside the tradition, transformed knowers disagree.

In Nyāya, which I have read more of, a cognition is confirmed from outside, ultimately by whether acting on it succeeds. Is there any analogue here? Or is the honest answer that mystical knowledge, if it exists, simply cannot be publicly checked, and that this is a feature rather than a bug?