Method transpose

tests/python/disco/test_session.py:234–238

(A: T.Buffer((8, 16), "float32"), B: T.Buffer((16, 8), "float32"))

    class TestMod:
        @T.prim_func(s_tir=True)
        def transpose(A: T.Buffer((8, 16), "float32"), B: T.Buffer((16, 8), "float32")):
            for i, j in T.grid(16, 8):
                with T.sblock("transpose"):
                    vi, vj = T.axis.remap("SS", [i, j])
                    B[vi, vj] = A[vj, vi]

        @R.function
        def main(A: R.Tensor((8, 16), dtype="float32")) -> R.Tensor((16, 8), dtype="float32"):