update rel cypher queries for migrated kind nodes

[ajtmccarty]

a test depends on this PR in the sdk opsmill/infrahub-sdk-python#403
will update the commit pointer once that PR is merged

updates to several Relationship-related cypher queries to correctly handle nodes that have had their kind or inheritance updated (in which case multiple nodes exist with the same UUID)

basically a more generic version of #6421

[codspeed-hq]

CodSpeed Performance Report

Merging #6433 will not alter performance

_{Comparing ajtm-05082025-rel-query-updates (849c1ce) with stable (94beb5d)}

Summary

✅ 10 untouched benchmarks

[dgarros]

how is this logic different from the branch filter we are usually using ?
also would it make sense to centralize this logic in order to reuse it, similar to what we are doing with the branch filters ?

[ajtmccarty]

the problem with the current logic is that if there are multiple edges returned by this part of the subquery

MATCH (rl)<-[active_edge:<EDGE_TYPE>]-(n)
WHERE %(active_rel_filter)s AND active_edge.status = "active"

then this will always return the active edge, even a more recent deleted edge exists

as for centralizing this logic somewhere, I think that might be valuable as part of the cypher query engine we've discussed before, but I don't think there's a good place for it right now