new(tests): EOF - EIP-3540: out of order container section

[chfast]

🗒️ Description

Test containers section being out of order in the header and/or body.
This extends and follows the convention of the test_section_order()
for the optional container section.

This also converts EOFTests/efValidation/EOF1_section_order_.json.

✅ Checklist

• All: Set appropriate labels for the changes.
• All: Considered squashing commits to improve commit history.
• All: Added an entry to CHANGELOG.md.
• All: Considered updating the online docs in the ./docs/ directory.
• Tests: All converted JSON/YML tests from ethereum/tests have been added to converted-ethereum-tests.txt.
• Tests: A PR with removal of converted JSON/YML tests from ethereum/tests have been opened.
• Tests: Included the type and version of evm t8n tool used to locally execute test cases: e.g., ref with commit hash or geth 1.13.1-stable-3f40e65.
• Tests: Ran mkdocs serve locally and verified the auto-generated docs for new tests in the Test Case Reference are correctly formatted.

[chfast]

The coverage difference is the following:

This is because we "fixed" the original tests by avoiding unrelated errors: unreferenced subcontainers or subcontainers of wrong kind (LBC). The original tests had it wrong because the mentioned errors were introduced after the tests were created.
I believe the converted version of the tests is better because it exercises the happy path of subcontainer validation (GBC).

[marioevz]

LGTM. Thanks a lot for the coverage explanation!

[marioevz]

This is because we "fixed" the original tests by avoiding unrelated errors: unreferenced subcontainers or subcontainers of wrong kind (LBC). The original tests had it wrong because the mentioned errors were introduced after the tests were created. I believe the converted version of the tests is better because it exercises the happy path of subcontainer validation (GBC).

Yes, I guess this will only be an issue for converted tests for which the spec has changed. Thanks for all the details on this, very helpful indeed.

[winsvega]

You see the coverage script is helpful even if it gets red.

I also found a few issues in my tests because of the cocerage report.

[chfast]

You see the coverage script is helpful even if it gets red.

I also found a few issues in my tests because of the cocerage report.

Yes, it is very helpful. I had to go back and add some missing test context multiple times. However, sometimes it may be considered unnecessary (https://github.com/ethereum/execution-spec-tests/pull/741/files#diff-ca4727af81398c427b5141ac76705400e44ba6d1afc25500f82032e1f8904756R227).