`ppi_mean_ci` does not support multidimensional inference problems

[PierreBoyeau]

The ppi_mean_ci function does not handle well cases where labelled/unlabelled datapoints are multidimensional.

The observation weights do not broadcast well with the Y matrix.
For instance, here, weights of shape (n_obs,), can't be multiplied with Y, when the latter are of shape (n_obs, d).