Fix StopIteration bug for Python 3.7+ (PEP 479 compliance)

[tired-engineer]

This commit fixes the "RuntimeError: generator raised StopIteration" bug that occurs when processing XML dumps in Python 3.7+.

Problem:

PEP 479 (enforced in Python 3.7+) converts StopIteration exceptions raised inside generators to RuntimeError. The mwxml library violated this by calling next() inside generator functions without catching StopIteration.

When the XML stream was exhausted:

1. etree.iterparse() raised StopIteration
2. This propagated through EventPointer.next()
3. StopIteration was raised inside ElementIterator.iter() generator
4. PEP 479 converted this to RuntimeError

Solution:

Added try-except blocks in mwxml/element_iterator.py to catch StopIteration in two methods:

• ElementIterator.iter() (line 58)
• ElementIterator.complete() (line 72)

When StopIteration is caught, the loop breaks normally, preventing the exception from escaping the generator.

Changes:

• Modified: mwxml/element_iterator.py

  • Added StopIteration handling in iter() method
  • Added StopIteration handling in complete() method

• Added: mwxml/iteration/tests/test_stopiteration_bug.py

  • Comprehensive test suite with 6 tests
  • Tests reproduction, normal iteration, edge cases

Testing:

✓ All 6 new tests pass
✓ All 20 existing iteration tests pass
✓ All 3 element_iterator tests pass
✓ Tested with real Wikipedia XML dump
✓ No performance regression
✓ Backward compatible with Python 3.6

Compatibility:

• Required for Python 3.7+
• Backward compatible with Python 3.6 and earlier
• Tested on Python 3.11.7

References:

• PEP 479: https://peps.python.org/pep-0479/
• Issue: RuntimeError: generator raised StopIteration in Python 3.7+