How would they unbrick a device? By definition, to "brick" is to render the device inoperable.
A friend of mine, in the course of his work, bricked a BMW [1]. It took an engineer, flown in [2] to fix the issue.
[1] He worked at a company making automobile diagnostic tools for mechanics. He was reflashing the BMW when the power went out. That scrambled the on-board computer system enough to brick the car.
[2] South Florida (Miami/Ft. Lauderdale area) I'm sure the German engineer wasn't all that upset.
To misquote The Princess Bride: "This device isn't bricked. It's just mostly bricked. There's a difference."
The device is still there, and it will talk, there's just no matching driver in the system for it now.
So you add PID=0x0000 to your existing driver (I think this is just an edit to the INF file on Windows) and sniff the device to see if it's one you've likely (semi) bricked, and make the attempt. You own the VID, so this operation should be safe in your namespace.
Brickness is a measure of the user's ability and motivation to restore a device. I've worked on systems where you have to unsolder really tiny devices in order to unbrick. It's a continuum of inconvenience, really. To most users, the evil FTDI driver made a brick because it was beyond their skill to do the repair (hey, you can write a driver for PID=0x0000 any time you want to, oaky?)
A friend of mine, in the course of his work, bricked a BMW [1]. It took an engineer, flown in [2] to fix the issue.
[1] He worked at a company making automobile diagnostic tools for mechanics. He was reflashing the BMW when the power went out. That scrambled the on-board computer system enough to brick the car.
[2] South Florida (Miami/Ft. Lauderdale area) I'm sure the German engineer wasn't all that upset.