[MXNET-1414]fix gradient bug of boolean mask

[wkcn]

Description

Related Issue: #15172

Hi, there.
The operator mx.nd.contrib.boolean_mask gives the wrong gradient.
I have fixed it and add a unittest.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

• The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage:
• Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
• Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
• Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
• Code is well-documented:
• For user-facing API changes, API doc string has been updated.
• For new C++ functions in header files, their functionalities and arguments are documented.
• For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
• Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
• To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

• Support Gradient Addition for boolean mask operator
• Fix the gradient bug
• Add a related unittest

Comment

In the CPU backward of boolean mask, it uses BooleanMaskBackwardCPUWriteKernel kernel to accelerate when req is WriteTo.

[wkcn]

Hi, @szha, @eric-haibin-lin . Could you please help review this PR?
It is related to the feature of MXNet 1.5.0.
Thank you!

[junrushao]

It looks good on my side. @haojin2 Could you also take a look? Thanks!

[wkcn]

I have merged this PR. Thank you! @junrushao1994 @haojin2