[API] Multi-level Stage Access

[zzzDavid]

• Related test: mlir/test_schedule_stream.py::test_dataflow_primitive

The kernel is defined as:

    def kernel(A, B):
        C = hcl.compute((10, 32), lambda *args : 0, "C")
        D = hcl.compute(C.shape, lambda *args: 0, "D")
        with hcl.for_(0, 10, tag="Super") as j:
            hcl.update(D, lambda *args: j*A[args] + B[args], name="update_D")
            hcl.update(C, lambda *args: A[args] + j*D[args], name="update_C")
        return C

We need to support multi-level stage accessing such as kernel.Super.update_D

[chhzh123]

Closed the issue since we no longer have the "stage" concept.