Possible bug in `poisson_weak`

[gaurav-arya]

In the poisson_weak implementation:

https://github.com/probcomp/adev/blob/57452f319d8edf098e9d6e8235c6f6864d434f90/src/Numeric/ADEV/Diff.hs#L126-L133

it looks like the derivative contribution from the continuation might be missing from the returned dual number. i.e. something like grad * rate' + tangent y_neg rather than just grad * rate'.