Make LLL reduction for integer matrix kernel optional

[gh-RikVoorhaar]

In the _right_kernel_matrix method of matrix_integer_dense there is a choice of 3 algorithms. With choice algorithm='pari' it calls the matkerint routine of PARI. According to the documentation this "gives an LLL-reduced ℤ-basis". The matker(x, flag = 1) routine omits the LLL-reduction, and for large matrices is much faster. It seems that the padic and flint algorithms also compute this LLL-reduced form.

In some applications this LLL reduction is not necessary.

I recommend documenting this fact and making LLL-reduction optional (but perhaps default).

In my particular application all three algorithms about equally fast, but directly calling

M.__pari__().matker(flag=1).mattranspose().sage()

gives a speedup of several orders of magnitude for large matrices. (In my case sparse matrices of shape around 2000x1000.)

[mkoeppe]

Batch modifying tickets that will likely not be ready for 9.1, based on a review of the ticket title, branch/review status, and last modification date.

[mkoeppe]

Setting new milestone based on a cursory review of ticket status, priority, and last modification date.