Fix segfault when using ALL() for rel comparison

[mdbenito]

This is a followup to #608 , completing the porting of the fixes in predictable-labs/ryugraph#2.

Instead of the custom data in the original PR I used the tinysnb dataset.

I also added another change from that PR, defensively avoiding an assert in debug builds in NodeOrRelExpression::addPropertyExpression

[adsharma]

[2026-06-22 14:55:30.271] [info] DEBUG LOG: RecursivePathWithAllAndRelComparison
[2026-06-22 14:55:30.271] [info] QUERY: MATCH (a:person)-[t:knows]->(d:person) WHERE a.ID <> d.ID WITH a, d, t MATCH p = (a)-[:knows*1..3]->(d) WHERE length(p) >= 2 AND ALL (r IN relationships(p) WHERE r <> t) WITH DISTINCT t RETURN count(t);
/home/runner/work/ladybug/ladybug/test/test_runner/test_runner.cpp:200: Failure
Value of: queryResult->isSuccess()
  Actual: false
Expected: true
Unexpected error for query: Assertion failed in file "/home/runner/work/ladybug/ladybug/src/include/common/cast.h" on line 30: false

[mdbenito]

Yes, sorry, I thought I had rebased on the correct branch. Part of the fix was on my previous PR, so this one could not pass the test. Now it should be ok.

[adsharma]

Something like:

https://gist.github.com/adsharma/9c189d8bec7df072cd3cae3ba30acd6d

has the same effect but less repeated code.

[mdbenito]

Actually there is no repetition, it's github being confused by the rebasing. This PR adds two lines. Let me rebase again on main

[adsharma]

Yes, the duplicate code already went into main #608, but could be optimized via the change above, since the fall through code passes tests even after eliminating the

if (child->expressionType == ExpressionType::PATTERN)

check inside the loop.

[mdbenito]

Ahh, I see what you mean. Just check if it's a pattern outside so we don't repeat what comes after the loop. Done!