Standard error of the overdispersion estimates

Dear Constantin,

I was wondering when estimating the overdispersion by `glmGamPoi::glm_gp()` function, can we extract the standard error of the estimates?

Also, when the algorithm shrinks the dispersion, would that impact the standard error?

Thank you!

Cheers,
Angli

Start at the glmGamPoi::glm_gp() entry point and its dispersion-estimation documentation. Trace how overdispersion estimates are produced and how shrinkage is applied, then define what standard-error output should mean. Done requires a decided API or documented answer, with validation of the estimate and shrinkage behavior.

Written by the indexing model from the issue text.