[Bug tree-optimization/93771] SLP produces VEC_PERM when should have used vector generation

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93771

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
If t[3] was an unrelated load, then SLP does the correct thing.
E.g.
void f(double *a, double *t, double *t0, double *d)
{
  double t1 = t[0] + d[0];
  double t2 = t0[0] + d[1];
  a[0] = t1;
  a[1] = t2;
}

Note this was reduced from PR 93720 comment #2.