Change the elementwise broadcasting contract from graph to kernel (#3894

)

Summary:
Pull Request resolved: #3894

Currently, there is a graph level pass to handle limited broadcasting of elementwise ops if the input tensors are not of the same size.

We move this responsibility down to the kernels with this diff, which is how ET and the portable ops do it. Ops of this kind are only `add`, `sub`, `mul` and `div` for now, but there will be more.

We retain the implementations for the reference kernels, because we want to avoid linking the portable ops directly, which takes forever at compile time. We can also use a much smaller set of types (basically only `float`).

We can remove a hack in the RNNT Joiner with this change, and run it natively. It takes a huge hit in performance, which will be fixed by getting broadcast-friendly kernels from Cadence.

We finally remove the binop tests in `test_aten_ops.py`, which were also using strange types and had been on the chopping block for a while.
We also remove the rSubScalar test, since we don't trace to rsub anymore.

Reviewed By: dulinriley

Differential Revision: D58207691

fbshipit-source-id: 12ec6df3b37eaea7ec1a7327c39a0a659bfaf1c0

kernels/portable/cpu/util/targets.bzl

@@ -73,7 +73,7 @@ def define_common_targets():
             "//executorch/runtime/kernel:kernel_includes",
             "//executorch/runtime/core/exec_aten/util:tensor_util",
         ],
-        visibility = ["//executorch/kernels/portable/cpu/...", "//executorch/kernels/optimized/cpu/..."],
+        visibility = ["//executorch/kernels/portable/cpu/...", "//executorch/kernels/optimized/cpu/...", "@EXECUTORCH_CLIENTS"],
     )
 
     runtime.cxx_library(