Poor error reporting for lockdown kernel mode failures (cannot set CC mode when secure boot is enabled)

It appears these tools do not work or gracefully report a useful error message if secure boot and the kernel lockdown=integrity option is enabled.

The relevant error output from
sudo ./nvidia_gpu_tools.py --gpu-bdf 0000:18 --query-cc-mode

is the following:

2025-08-23,22:03:14.042 ERROR    Device /sys/bus/pci/devices/0000:18:00.0 broken [Errno 1] Operation not permitted
Topo:
  Intel root port 0000:15:01.0
   PciBridge 0000:16:00.0 c030:1000
    PciBridge 0000:17:00.0 c030:1000
     GPU 0000:18:00.0 ? 0x2335 BAR0 0x19c042000000
     GPU 0000:18:00.0 [broken, cfg space working 1 bars configured 1]
2025-08-23,22:03:14.043 INFO     Selected GPU 0000:18:00.0 [broken, cfg space working 1 bars configured 1]
Traceback (most recent call last):

If you look in kernel logs, you can also find this:
Lockdown: python3: direct PCI access is restricted; see man kernel_lockdown.7

Start with nvidia_gpu_tools.py and reproduce the failure using --gpu-bdf 0000:18 --query-cc-mode with secure boot and kernel lockdown=integrity enabled. Inspect how the [Errno 1] Operation not permitted result is handled, then verify that the tool reports a useful lockdown-related error instead of continuing with a broken device and traceback.