FPE-7: document M8 as structural obstruction, M9 as open selector gap Corrects (A1)/(A2) attribution

[Copilot]

The self-consistency map 𝓕's well-definedness carried two open structural gaps: (M8) no construction ever showed a spacetime fixed point is equivalent to a fixed point of Cauchy data on a chosen Σ, and (M9) 𝓕's single-valuedness relied on an unproven "unique perturbation" state-selection claim. The (A1)/(A2) assumption tags also over-attributed continuity properties to Hollands–Wald that their cited result doesn't establish.

Changes

• M8 — documented as a structural obstruction. Tied directly to the already-closed FPE-4/M3 finding: the same missing spatial reduction of the Lorentzian, causal-past-supported response kernel that blocked M3 also blocks the spacetime-fixed-point ↔ Σ-data-fixed-point equivalence at the generality of (A1)–(A6). Any repair is foliation-restricted and local in time, consistent with FPE-4's existing Outcome B — not reopened or re-litigated here.
• M9 — documented as an open selector gap. Hadamard regularity specifies an admissible class of states, not a unique selector g ↦ |Ψ_g⟩. Engages informs-issue [informs-issue] arXiv:2606.22767 — Philosophical challenge to Hadamard state foundations; informs FPE-7 state-selection prescription #140's critique of Hadamard-condition necessity (exploratory tier, unverified beyond abstract): compatible with treating Hadamard regularity as defining an admissible class rather than a unique state, but doesn't by itself close the gap. Closing M9 requires either a verified selector (symmetry, adiabatic branch, variational principle) with proven continuity, or a Kakutani-style reformulation with hypotheses checked.
• (A1)/(A2) attribution corrected. Hollands–Wald establishes locally covariant algebraic Wick-polynomial/time-ordered-product renormalization, not the paper's assumed H^s → H^{s-2} continuity for a chosen family of states. Replaced the "Established" tags with an accurate "assumed, not established by the cited results" framing.
• Order-reduction paragraph softened (§sec:banach) to flag that the "adiabatic vacuum state at each order" prescription is exactly what M9 leaves open, rather than an established theorem.

Both gaps remain at Sketch level — no Rigorous claim is made for either the obstruction statement or a resolution.