WIP: Update dockerisation

[parithosh]

This PR adds a script to run the tests inside a Docker container. Ideally we can move to using a containerized environment for all the tests in the future, but the script is a good first step at gathering user feedback.

The PR adds a standalone Dockerfile, this file just sets up the dependencies required to run the tool and can be used independently. The docker image can be locally built with:

• docker build ./ -t $IMAGE_NAME -f ./docker/Dockerfile will build the docker container
• docker run -it $IMAGE_NAME /bin/sh will give you a shell inside the docker container to manually run any tests
• docker run $IMAGE_NAME make citest will run the make citest command inside the docker container
• Ideally manual running of docker containers is for advanced users, we recommend the script for most users

There is additionally a scripts/build_run_docker_tests.sh script which covers most regular usecases. The script allows the user to configure the fork(altair/bellatrix/capella..), image_name(specifies the container to use), number of cores, preset type(mainnet/minimal) and test all forks flags. Ideally this is the main way that users interact with the spec tests, instead of running it locally.

E.g:

• ./build_run_test.sh --p mainnet --n 16 will run the mainnet preset tests with 16 threads
• ./build_run_test.sh --a will run all the tests across all the forks
• ./build_run_test.sh --f deneb --n 16 will only run deneb tests on 16 threads

Results are always placed in a folder called ./testResults. The results are .xml files and contain the fork they represent and the date/time they were run at.

[mrthankyou]

@parithosh,

Do you need any additional help on this PR? This is semi-blocking for me as I would like to write some tests for consensus-spec and this would make my work much easier :).

[parithosh]

Sorry I haven't had much time to look into the topic since the initial commits. I'll try and assign some time next week to close up the topic.

[mrthankyou]

Happy to help and save you time any way I can. Just let me know!

[parithosh]

cc @hwwhww @dapplion @mrthankyou

Sorry for the delay on this PR, let me know what you think and how it works for you.

[hwwhww]

@parithosh Tested locally and LGTM! 👏

p.s. It's probably better to squash merge it

[parithosh]

would it make sense to add a small usage/installation guide in the main README as well?

[hwwhww]

@parithosh

Maybe add tutorial in a docker/README.md file and add a link in the main home README.md?