fix custom op for backward compatibility

[ZiyueHuang]

Description

The default implementation infer_storage_type_backward in custom op has some problem.

Before this PR, the LR in sparse example is broken due to the custom op, and the added test will throw err message,

Error in mult.infer_type: Traceback (most recent call last):
  File "/home/hanfeng/zyh/mxnet/python/mxnet/operator.py", line 737, in infer_storage_type_backward_entry
    "stypes, got %d."%(total_outputs, len(ostype))
AssertionError: InferStorageTypeBackward Error: expecting 2 entries in returned output stypes, got 1.

[10:27:12] /home/hanfeng/zyh/mxnet/dmlc-core/include/dmlc/./logging.h:308: [10:27:12] src/operator/custom/custom.cc:383: Check failed: reinterpret_cast<CustomOpBackwardInferStorageTypeFunc>( params.info->callbacks[kCustomOpPropBackwardInferStorageType])( stypes.size(), stypes.data(), params.info->contexts[kCustomOpPropBackwardInferStorageType])

cc @eric-haibin-lin @anirudh2290

Checklist

Essentials

• Passed code style checking (make lint)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage
• For user-facing API changes, API doc string has been updated. For new C++ functions in header files, their functionalities and arguments are well-documented.
• To my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

• Feature1, tests, (and when applicable, API doc)
• Feature2, tests, (and when applicable, API doc)

Comments

• If this change is a backward incompatible change, why must this change be made.
• Interesting edge cases to note here

[eric-haibin-lin]

@anirudh2290

[anirudh2290]

Thank you for fixing this and adding the tests! Yes, since the output stypes which is the in_grads would depend on arguments length and not output length. Can you please let me know what LR example you are referring to?

[anirudh2290]

Please remove whitespaces here and above.

[ZiyueHuang]

@anirudh2290 Thanks for your comments. LR example is the linear_classification.py in example/sparse.

[anirudh2290]

Found an issue with custom_op when used with need_top_grad=False. Fixing here: #8725

[piiswrong]

Closing this because we are reverting sparse support for custom op