Consider making neoplasm and syndrome disjoint classes

[cmungall]

Such that for example Lynch syndrome would not be a subclass of neoplasm but hereditary neoplasm-disposing syndromes but would instead connect via some other relationship

cc @dosumis @mellybelly @matentzn