Function cross_covariance 

fn cross_covariance(
    model: &Model<'_, '_>,
    x: &Xccy,
    s: f64,
    t: f64,
) -> ([[f64; 7]; 7], (f64, f64))

The covariance of (x, y, ∫x, ∫y, x_f, ∫x_f, σ_S·W) over [s, t] given the state at s, with the means the collateral factor’s quanto drift −ρ_fS·σ_f·σ_S adds to x_f and ∫x_f.

Every entry is one kernel over v = t − r on the pieces where both volatilities are flat: with E(k) = ∫e^{−kv}dv, a level with a level E(k_i+k_j), a level with an integral (E(k_i) − E(k_i+k_j))/k_j, two integrals (Δ − E(k_i) − E(k_j) + E(k_i+k_j))/(k_i·k_j); the FX innovation is a level of mean reversion zero.