[23.2] Yaml nested assertions: fix parsing

[bernt-matthias]

Seems that the recursive call was forgotten.

How to test the changes?

(Select all options that apply)

• I've included appropriate automated tests.
• This is a refactoring of components with existing test coverage.
• Instructions for manual testing are as follows:
  1. [add testing steps and prerequisites here if you didn't write automated tests covering all your changes]

License

• I agree to license these and all my past contributions to the core galaxy codebase under the MIT license.

[mvdbeek]

I guess that's what your comment about asserts was about ? It does feel more consistent for that to be

asserts:
  - has_text:
    text: thing

?

[bernt-matthias]

I used the syntax from the other examples in the file

I'm still struggling with yaml, but it seems that

yaml.load("""
asserts:
  - has_text:
    text: thing
"""
)

gives `{'asserts': [{'has_text': None, 'text': 'thing'}]}`

while

```python
yaml.load("""
asserts:
  - has_text:
      text: thing
"""
)

gives {'asserts': [{'has_text': {'text': 'thing'}}]}. Not sure if both versions are parsed at the moment.

Not sure if the former even works.

[mvdbeek]

Sorry, typing without checking, it should be

asserts:
  - has_text:
      text: thing

[bernt-matthias]

Yes this is what I meant with my comment at the planemo issue

Not sure, but easy to implement I guess. But I can clearly live with the current state. The only question would be if it would simplify anything if we use asserts?

[mvdbeek]

Yes, then we can re-use models and there's no question what you have to use IMO - it's always asserts.

[bernt-matthias]

Then remove children completely, or leave it as falll back? I guess since it seems that this was never used, we may just change it?

[mvdbeek]

Leave it as a fallback but don't document it would be my suggestion ?

[github-actions]

This PR was merged without a "kind/" label, please correct.