Don't pull docker image in AWS Codebuild

[Fisher-Wang]

Sometimes the aws CI check would encounter the following error

#2 [metasim internal] load metadata for docker.io/nvidia/cuda:11.8.0-runtime-ubuntu22.04
#2 ERROR: failed to copy: httpReadSeeker: failed open: unexpected status code https://registry-1.docker.io/v2/nvidia/cuda/manifests/sha256:7c8e5ee1e8531f0d19f8d7e78c5b7efa9c455697fa96e62f083f361d6842e452: 429 Too Many Requests - Server message: toomanyrequests: You have reached your unauthenticated pull rate limit. https://www.docker.com/increase-rate-limit
failed to solve: nvidia/cuda:11.8.0-runtime-ubuntu22.04: failed to resolve source metadata for docker.io/nvidia/cuda:11.8.0-runtime-ubuntu22.04: failed to copy: httpReadSeeker: failed open: unexpected status code https://registry-1.docker.io/v2/nvidia/cuda/manifests/sha256:7c8e5ee1e8531f0d19f8d7e78c5b7efa9c455697fa96e62f083f361d6842e452: 429 Too Many Requests - Server message: toomanyrequests: You have reached your unauthenticated pull rate limit. https://www.docker.com/increase-rate-limit
------
 > [metasim internal] load metadata for docker.io/nvidia/cuda:11.8.0-runtime-ubuntu22.04:
------
Error: Process completed with exit code 17.

it is about unauthenticated Docker Hub rate limit. To fix it, follow this answer

[Fisher-Wang]

AWS seems not an ideal choice for simulator CI. It turns out the core problem is that its GPU is not ready for both torch computing and simulation computing. See #178 for more details. #70 is also a trial.