Ya unfortunately I don't have any special inside information, couldn't get any reply on the DirectX boards and like you was not able to find any information online. I just tried everything with tracing and related functions to try to make sense of it.
When I get back in town I'll break down how I got it working with warp device and where to make the proper calls. I was able to set the trace, render and disassemble executed code and get valid register values and think I have enough information to map from the assembly to my HLSL shader code but just need hardware support because debugging this in warp is almost as bad as using ref device. I had to put it aside to get some real progress done until I know more about it's support.
Still looking to get information from AMD and NVidia on why this extremely powerful DX 11.1 feature in particular is not supported in hardware yet.
Cheers