[MXNET-1253] [WIP] Fix large array issue in topk operator

[apeforest]

Description

This is a second batch of fixes for several sorting operators.

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

• topk
• test_large_array.py

Comments

• Removed dtype for topk operator for indices. It seems unnecessary (and confusing) to have float32 as a type for indices and not clear why user would ever want to define it. This may cause backward compatibility issue only if users set dtype for return indices of a topK operator.

[Roshrini]

@apeforest Thanks for working on this.
@mxnet-label-bot Add [pr-work-in-progress, Operator]

[TaoLv]

Looks like an incompatible change.

[zheng-da]

agree. could you please revert the change?

[Roshrini]

@apeforest Can you address comments and rebase this PR? Thanks

[stu1130]

@apeforest ping again for the update

[apeforest]

Got fully occupied with another project. I will resume this PR when I get some slack next week.

[ankkhedia]

@apeforest Did you get a chance to work on this PR?
#13627 depends on this fix.

[pinaraws]

@apeforest Any updates?

[piyushghai]

@apeforest Can you rebase with the current master branch and look into the CI failures ?

[Roshrini]

@apeforest Any update on this PR?

[pinaraws]

@apeforest Any update on this PR?

[apeforest]

@access2rohit is working on this in a new PR. Closing this one.