theorem gibbs_static_vri [NeZero J] (ε x : Fin J → ℝ) (T : ℝ)
(_hT : 0 < T) :
let P := gibbsProb ε x T 0
let μ := ∑ j, x j * P j
let σ_sq := (∑ j, x j ^ 2 * P j) - μ ^ 2
-- The VRI: σ² = T · χ, expressed as σ² = Σ P_j · x_j · (x_j - μ)
-- (the right side equals T · χ because χ = (1/T) · Σ P_j · x_j · (x_j - μ))
σ_sq = ∑ j, P j * x j * (x j - μ) :=
(algebraic_vri_core (gibbsProb ε x T 0) x (gibbsProb_sum_one ε x T 0)).symm### The Variance-Response Identity