Use xarray-dataset to save all data

[ahartikainen]

I think we should store all the data from the sampling straight to xarray.Dataset. This way the results are in a compact place and we have a good way to access it.

Not sure should also store functions there or only the data.

See. e.g. arviz-devs/arviz#97

[ahartikainen]

Or maybe we could still use a dictionary, but save with the same format

{'theta' : arr}
# arr is ndarray
arr.shape = (chain, draw, *shape)

Or

{'theta' : arr}
# arr is List of ndarray
arr[i].shape = (draw, *shape)

Or

[{'theta' : arr}]
# arr is ndarray
arr.shape = (draw, *shape)

Then we can easily transform to xarray.Dataset.

Should we try and kick some tyres, now the version is 3.0.x alpha?

cc. @mkcor

[ahartikainen]

I think with arviz.from_pystan works, so we can keep Fit object as a simple numpy object.