No, I'm really not interested. The card functions in what is basically a hardware RAID mode. We already knew that. The card's firmware is presenting an emulated disk to the host that happens to be a direct mapping of the drive, but it is still in the data path and providing its own representation.
What happens when a disk experiences an error? What happens when there's an unexpected interaction like a hot swap event? There are numerous questions, including unforeseen ones, whose answers we do not know.
You want "factual data"? What the hell is that? You expect someone else to prove that using an untested controller is bad? It doesn't work like that. The way it works is that we periodically see people wander on in here, with various janky hardware configurations, complaining that they're unable to boot their NAS because "something happened." Then we look at it, find out that there's some sort of failure of the underlying disks that got hidden from ZFS because of the controller, and then we all go "awwww you lost redundancy and now your pool is toast."
"But I want factual data" is the cry of someone who is intent on using their potentially risky hardware unless someone can come up with an example to demonstrate that it is pointing a gun at your foot and pulling the trigger. It doesn't work that way. We work based on the things we know to work swimmingly well. For example, current Intel SATA ports and 6Gbps LSI ports both have a billion hours of problem-free operation under their belts with FreeNAS with a huge base of installed users. We know these things work very well.
We *assume* you come to FreeNAS because you want your data protected securely. If that's an invalid assumption, by all means, go off and do whatever you want.