ENH: broadcast lstsq

Many `linalg` functions are already broadcastable, `lstsq` isn't.

A workaround is via `svd` which is already broadcasted:
```python
u, s, v = numpy.linalg.svd(A, full_matrices=False)
uTb = numpy.einsum('ijk,ij->ik', u, b)
xx = numpy.einsum('ijk, ij->ik', v, uTb / s)
```