Gpjax

Fix stability in Matérn kernels.

What's Changed
* Transformations by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/109
* Natgrads by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/90
* Fix bug in Matern12 kernel by thomaspinder in https://github.com/thomaspinder/GPJax/pull/119
* Intro to GP notebooks by thomaspinder in https://github.com/thomaspinder/GPJax/pull/117
* Add verbose option. by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/116
* Numpyro by thomaspinder in https://github.com/thomaspinder/GPJax/pull/122
* Distrax reversion by thomaspinder in https://github.com/thomaspinder/GPJax/pull/125
* Kernel compute by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/120
* Prevent f64 default by thomaspinder in https://github.com/thomaspinder/GPJax/pull/129
* Update w/ Dan comments by thomaspinder in https://github.com/thomaspinder/GPJax/pull/130
* Cleanup reqs by thomaspinder in https://github.com/thomaspinder/GPJax/pull/131
* Update docs by thomaspinder in https://github.com/thomaspinder/GPJax/pull/137
* Improve readability and add comments. by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/138
* Move params to the first slot of each function, class, etc. by Daniel-Dodd in https://github.com/thomaspinder/GPJax/pull/139
* V0.5 update by thomaspinder in https://github.com/thomaspinder/GPJax/pull/123


**Full Changelog**: https://github.com/thomaspinder/GPJax/compare/v0.4.13...v0.5

What's Changed
* Refactor docs as markdown documents by thomaspinder in https://github.com/thomaspinder/GPJax/pull/107
* Refactor JaxTyping to be compatible with v0.0.2
* Pin pypa-publish workflow

**Full Changelog**: https://github.com/thomaspinder/GPJax/compare/v0.4.12...v0.4.13

This minor release resolves the issue surrounding Distrax transformations.

A PRNGKey can now be passed to the initialisation function for reproducible parameter initialisation when parameters are stochastic e.g., RFFs. Further, the return argument of `initialise` and `fit` is a dataclass than bundles up the constituent quantities.

Minor change that fixes a bug in conjugate regression models where the marginal log-likelihood is evaluated on a single datapoint.