dmnorm implementation should invserve the precision matrix

[sunxd3]

In

JuliaBUGS.jl/src/BUGSPrimitives/distributions.jl

Lines 385 to 399 in adfa8ab

	"""
	dmnorm(μ::Vector, T::Matrix)
	Return a [Multivariate Normal](https://juliastats.org/Distributions.jl/latest/multivariate/#Distributions.MvNormal)
	distribution object with mean vector `μ` and precision matrix `T`.
	The mathematical form of the PDF for a Multivariate Normal distribution in the BUGS family of softwares is given by:
	```math
	p(x|μ,T) = (2π)^{-k/2} |T|^{1/2} e^{-1/2 (x-μ)' T (x-μ)}
	```
	"""
	function dmnorm(μ::Vector, T::Matrix)
	return MvNormal(μ, T)
	end

consider https://juliastats.org/Distributions.jl/latest/multivariate/#Distributions.AbstractMvNormal and https://www.multibugs.org/documentation/latest/Distributions.html#MultivariateNormal

The precision matrix should be inverted before creating the MvNormal distribution

[sunxd3]

fixed by #106