Cholesky in-place inversion high memory use

[jbncode]

The workspace for the Cholesky inversion functionfaer::linalg::cholesky::llt::inverse::invert_lower_in_place as computed by invert_lower_in_place_req is equal to the size of the original matrix, doubling the memory requirements. Is this a bug? If not, is there a way to avoid the extra memory use and do a true in-place inversion?

[sarah-quinones]

why do you want to invert a matrix?

[jbncode]

For example to compute a covariance matrix.