Remove StatusUpdates that include a U2FDeviceInfo

[jschanck]

(This is conceptually part of #270, but I separated it out since it has side effects.)

The abstract FidoDevice trait depends on U2FDeviceInfo through

fn get_device_info(&self) -> U2FDeviceInfo;
fn set_device_info(&mut self, dev_info: U2FDeviceInfo);

The data in U2FDeviceInfo is specific to the HID transport, and it's somewhat awkward to include it in virtual devices.

Apart from checking for the CBOR capability in FidoDevice::init (which is probably not necessary), we only use get_device_info in status updates. Firefox doesn't do anything with these status updates, so I don't see any reason to keep them.

[mozkeeler]

LGTM!

[msirringhaus]

At least in the current draft of the about-page, I am relying on the DeviceAvailable/Unavailable messages (e.g. to determine when resetting a device, if all have been removed and only the one we want to reset has been re-inserted).
It is not using the dev-info, however (yet). If #217 would land, then we could use that to display the names of the token and the manufacturer, too (not strictly needed, but nice to have).
So I feel hesitant to remove it all, although I do agree that its a bit awkward in general and could use some refactoring.

[jschanck]

Your use of DeviceAvailable and DeviceUnavailable on the about page doesn't require a U2FDeviceInfo, so I would at least want to simplify the status updates. But I'd suggest that, rather than counting DeviceAdded and DeviceUnavailable messages, you add a StatusUpdate::ResetError(...) that tells the application what needs to be done to perform a reset.

To your second point, we could include a human-readable debug_info: String with StatusUpdate::InteractiveManagement and let each device type define what goes in it.