fix(fill,execute): register new label type markers to prevent warnings

[danceratopz]

🗒️ Description

Follow-up to #1220.

1. Register the new label type markers that get applied to tests to avoid pytest test session warnings, e.g.:

   ================================================== warnings summary ===================================================
   
   src/pytest_plugins/shared/helpers.py:37
     /home/dtopz/code/github/danceratopz/eest/src/pytest_plugins/shared/helpers.py:37: PytestUnknownMarkWarning: Unknown pytest.mark.state_test_from_eof_test - is this a typo?  You can register custom marks to avoid this warning - for details, see https://docs.pytest.org/en/stable/how-to/mark.html
       getattr(
   ...
   

   (can be reproduced via fill tests/osaka --fork=osaka).
2. Additionally adds a description field to LabeledExecuteFormat and LabeledFixtureFormat that can be used in the marker doc.

Fill

Command to view markers:

fill --markers

Relevant output:

@pytest.mark.blockchain_test: Tests that generate a blockchain test fixture.
@pytest.mark.blockchain_test_engine: Tests that generate a blockchain test fixture in Engine API format.
@pytest.mark.eof_test: Tests that generate an EOF test fixture.
@pytest.mark.state_test: Tests that generate a state test fixture.
@pytest.mark.transaction_test: Tests that generate a transaction test fixture.
@pytest.mark.eof_state_test: Tests that generate an EOF state_test.
@pytest.mark.eof_blockchain_test: Tests that generate an EOF blockchain_test.
@pytest.mark.eof_blockchain_test_engine: Tests that generate an EOF blockchain_test_engine.
@pytest.mark.state_test_from_eof_test: A state_test generated from an eof_test.
@pytest.mark.blockchain_test_from_eof_test: A blockchain_test generated from an eof_test.
@pytest.mark.blockchain_test_engine_from_eof_test: A blockchain_test_engine generated from an eof_test.
@pytest.mark.blockchain_test_from_state_test: A blockchain_test generated from a state_test
@pytest.mark.blockchain_test_engine_from_state_test: A blockchain_test_engine generated from a state_test
@pytest.mark.blockchain_test_engine_only: Only generate a blockchain test engine fixture
@pytest.mark.blockchain_test_only: Only generate a blockchain test fixture
@pytest.mark.eof_test_only: Only generate an EOF test fixture
@pytest.mark.eof_test_only: Only generate an EOF test fixture
@pytest.mark.state_test_only: Only generate a state test fixture

Execute

Command to view markers:

uv run execute remote --markers --rpc-seed-key=0xa --rpc-endpoint=foo --rpc-chain-id=1 --fork=cancun

Relevant output:

@pytest.mark.transaction_post: Simple transaction sending, then post-check after all transactions are included
@pytest.mark.eof_transaction_post: Tests that generate an EOF transaction_post.
@pytest.mark.transaction_post_from_eof_test: A transaction_post generated from an eof_test.
@pytest.mark.blockchain_test_engine_only: Only generate a blockchain test engine fixture
@pytest.mark.blockchain_test_only: Only generate a blockchain test fixture
@pytest.mark.eof_test_only: Only generate an EOF test fixture
@pytest.mark.eof_test_only: Only generate an EOF test fixture
@pytest.mark.state_test_only: Only generate a state test fixture

🔗 Related Issues

✅ Checklist

• All: Set appropriate labels for the changes.
• All: Considered squashing commits to improve commit history.
• All: Added an entry to CHANGELOG.md.

[danceratopz]

@marioevz please improve the descriptions, if you can! 🙂

[marioevz]

It seems like #1238 does something similar but doesn't have descriptions!

I'll merge 1238 first and then rebase this on top of that one to bring both together 👍

[marioevz]

Thanks for this! I like the descriptions more than in my previous implementation, it's nice that we enforce some thought on describing the labels/markers so we can guarantee quality.

I left the previous approach in #1238 instead of the one in this PR, but that's the only change I made. Will merge now.

[danceratopz]

Thanks for incorporating the descriptions @marioevz! I totally missed #1238, I definitely prefer #1238's approach for tracking the labels.