Use Constraints Instead of Plain Range Index for Node keys/ids

[zach-blumenfeld]

Regarding stuff like this in kg writer: https://github.com/neo4j/neo4j-graphrag-python/blob/main/src/neo4j_graphrag/experimental/components/kg_writer.py#L103

For unique node ids/keys used in loading, you want to use a Constraint: Either Key or Uniqueness constraint. Key is like uniqueness but also doesn't allow Nulls and is Enterprise only.