poisson_objective_grad: Poisson objective function gradient
In incidental: Implements Empirical Bayes Incidence Curves

This function computes the Poisson objective function (including regularizer) gradient.

1	poisson_objective_grad(beta, lam, reported, Q, lnPmat, regularization_order)

`beta`	spline parameter vector length num_params
`lam`	positive scalar regularization strength
`reported`	An integer vector of reported cases.
`Q`	spline basis matrix Tmod x num_params
`lnPmat`	matrix size Tobs x Tobs, log of make_likelihood_matrix
`regularization_order`	An integer (typically 0, 1, 2), indicating differencing order for L2 regularization of spline parameters. Default is 2 for second derivative penalty.