fix: prevent infinite loop in label_propagation when edge_count >= 2

[VictorECDSA]

The original while True loop could oscillate forever when entity pairs have edge_count >= 2. In that case candidate_rank > 1 causes branch A to lower a node's community ID, while the max() fallback in branch B raises it back up in the next iteration, creating a cycle that never satisfies no_change == True.

Fix: replace while True with a bounded for-loop using max_iterations = len(projection) * 10 + 10. The algorithm logic is unchanged -- the loop still breaks early on convergence. The community_map update is moved to before the break check (required by the for-loop structure to preserve the last iteration's result).

Also add tests/utils/maintenance/test_community_operations.py covering termination and correctness for the trigger condition (edge_count >= 2) and several other graph shapes.

Summary

Fix infinite loop in label_propagation() that caused build_communities() to hang indefinitely (100% CPU, no output, no error) when any entity pair in the graph had edge_count >= 2.

Type of Change

• Bug fix
• New feature
• Performance improvement
• Documentation/Tests

Objective

N/A (bug fix)

Testing

• Unit tests added/updated (tests/utils/maintenance/test_community_operations.py)
• Integration tests added/updated
• All existing tests pass

Without fix — pure Python reproduction (capped at 8 rounds):

round 1: {'A': 1, 'B': 0}
round 2: {'A': 0, 'B': 1}  <- back to initial state!
round 3: {'A': 1, 'B': 0}
...
❌ INFINITE LOOP: still oscillating after 8 rounds

With fix — same trigger graph:

Result clusters: [['A'], ['B']]
✅ Returned without infinite loop
✅ build_communities() returned successfully
✅ 4 community node(s) created in Neo4j

Breaking Changes

• This PR contains breaking changes

Checklist

• Code follows project style guidelines (make lint passes)
• Self-review completed
• Documentation updated where necessary
• No secrets or sensitive information committed

Related Issues

Closes #1355

[danielchalef]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[VictorECDSA]

I have read the CLA Document and I hereby sign the CLA