[Feature] Add oneDNN support for interleaved_matmul_selfatt_* operators (fp32/int8)

[bgawrych]

Description

This change adds oneDNN support for two operators:

• _contrib_interleaved_matmul_selfatt_qk
• _contrib_interleaved_matmul_selfatt_valatt

Both operators will be used when backend MKLDNN/MKLDNN_QUANTIZE will be chosen - there is no change in terms of performance between MKL fp32 vs. oneDNN fp32, but the main advantage is utilizing int8 data type

10 iterations of BERT-Large (gluon-nlp v0.10.x) [Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz]:

MKL implementation (fp32 as int8 is not supported):

oneDNN implementation (int8):

_contrib_interleaved_matmul_selfatt_qk => _sg_mkldnn_selfatt_qk
_contrib_interleaved_matmul_selfatt_qk => _sg_mkldnn_selfatt_valatt

We can observe that this change positively influenced other operators as there is less dequantization/quantization overhead and memory reorders

Great contribution of @grygielski to this change

Checklist

Essentials

• PR's title starts with a category (e.g. [BUGFIX], [MODEL], [TUTORIAL], [FEATURE], [DOC], etc)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage

[mxnet-bot]

Hey @bgawrych , Thanks for submitting the PR
All tests are already queued to run once. If tests fail, you can trigger one or more tests again with the following commands:

• To trigger all jobs: @mxnet-bot run ci [all]
• To trigger specific jobs: @mxnet-bot run ci [job1, job2]

---

CI supported jobs: [unix-cpu, clang, unix-gpu, windows-cpu, centos-gpu, website, centos-cpu, sanity, miscellaneous, windows-gpu, edge]

---

Note:
Only following 3 categories can trigger CI :PR Author, MXNet Committer, Jenkins Admin.
All CI tests must pass before the PR can be merged.

[bgawrych]

@mxnet-bot run ci [windows-cpu]

[mxnet-bot]

Jenkins CI successfully triggered : [windows-cpu]

[sfraczek]

lgtm

[bgawrych]

@szha Can we merge it?

[szha]

@bgawrych thanks for the contribution