CI tests failing when initializing containers

[FlorianReiss]

Since some time, multiple CI test are failing when initializing containers

Starting job container
  /usr/bin/docker pull nvidia/cuda:8.0-devel-ubuntu16.04
  Error response from daemon: manifest for nvidia/cuda:8.0-devel-ubuntu16.04 not found: manifest unknown: manifest unknown
  Warning: Docker pull failed with exit code 1, back off 4.685 seconds before retry.
  /usr/bin/docker pull nvidia/cuda:8.0-devel-ubuntu16.04
  Error response from daemon: manifest for nvidia/cuda:8.0-devel-ubuntu16.04 not found: manifest unknown: manifest unknown
  Warning: Docker pull failed with exit code 1, back off 5.4 seconds before retry.
  /usr/bin/docker pull nvidia/cuda:8.0-devel-ubuntu16.04
  Error response from daemon: manifest for nvidia/cuda:8.0-devel-ubuntu16.04 not found: manifest unknown: manifest unknown
  Error: Docker pull failed with exit code 1

Most likely the images we want to use are not supported any more https://gitlab.com/nvidia/container-images/cuda/blob/master/doc/unsupported-tags.md

From the list of containers it looks only CUDA12 would be available