The spinor components are antisymmetric by bipartite geometry within a session, but CompositeCausalSession combines sessions with bosonic coherent superposition. That’s a spin-statistics inconsistency — is it a bug in my composition rule or a real signature I could probe with H⁻ and helium?