Avoid unnecessary parsing conditions

[jkronegg]

🤔 What's changed?

Minor changes to avoid unnecessary stuff:

• avoid passing a default null value in AstNode.getSingle
• avoid duplicate condition checks
• avoid checking null value on elements that are never null (AstNode.getToken)
• avoid never used FeatureHeader and RuleHeader default value
• corrected bad lib scope for assertj-core (was compile-time)

⚡️ What's your motivation?

I didn't check the performance as theses are very small changes. But removing unnecessary code is never a bad thing...

🏷️ What kind of change is this?

• 🏦 Refactoring/debt/DX (improvement to code design, tooling, etc. without changing behaviour)

♻️ Anything particular you want feedback on?

I couldn't run make mostlyclean acceptance (these are many errors regarding newlines on Windows).

📋 Checklist:

• I agree to respect and uphold the Cucumber Community Code of Conduct
• I've changed the behaviour of the code
  • I have added/updated tests to cover my changes.
• My change requires a change to the documentation.
  • I have updated the documentation accordingly.
• Users should know about my change
  • I have added an entry to the "Unreleased" section of the CHANGELOG, linking to this pull request.

[mpkorstanje]

The AssertJ fix is good. I'll pull that out separately.

Beyond that I'm not sure how I feel about this change. Checking pre- and post-conditions catches mistakes in understanding early. The checks are also quite inexpensive. I don't think these milligrams of performance gain are worth the potential headache in the future..