Feedback request: Multivariate Probit Regression with GP

[avehtari]

@Matias_Guzman_Naranjo

vector diagSPD_EQ_2D(real sdgp, real lscale, real L, int M) {
    return sqrt((sdgp^2) * sqrt(2*pi()*lscale)^2 *
		exp(-0.5*(lscale*pi()/2/L)^2*linspaced_vector(M, 1, M)^2));
  }

I think sqrt(2*pi()*lscale)^2 should be 2*pi()*lscale^2 and the term inside exp seems to be missing *2 (compare how https://github.com/gabriuma/basis_functions_approach_to_GP/blob/master/multi_dimensional/simulated_dataset/model_BF_multi.stan#L23 has rho1^2*w1^2 + rho2^2*w2^2, and you have rho1=rho2 and w1=w2).