[Feature] Support MPS device.

[mzr1996]

Motivation

PyTorch 1.12.0 has supported using MPS to accelerate training on Mac.

Modification

Support auto select mps as device and use --device mps to specify device.

Checklist

Before PR:

• Pre-commit or other linting tools are used to fix the potential lint issues.
• Bug fixes are fully covered by unit tests, the case that causes the bug should be added in the unit tests.
• The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
• The documentation has been modified accordingly, like docstring or example tutorials.

After PR:

• If the modification has potential influence on downstream or other related projects, this PR should be tested with those projects, like MMDet or MMSeg.
• CLA has been signed and all committers have signed the CLA in this PR.

[codecov]

Codecov Report

Merging #894 (49c7f8c) into dev (812f3d4) will decrease coverage by 1.01%.
The diff coverage is 27.77%.

❗ Current head 49c7f8c differs from pull request most recent head 9a8c7f3. Consider uploading reports for the commit 9a8c7f3 to get more accurate results

@@            Coverage Diff             @@
##              dev     #894      +/-   ##
==========================================
- Coverage   85.45%   84.43%   -1.02%     
==========================================
  Files         132      134       +2     
  Lines        8750     8772      +22     
  Branches     1513     1517       +4     
==========================================
- Hits         7477     7407      -70     
- Misses       1050     1132      +82     
- Partials      223      233      +10     

Flag	Coverage Δ
unittests	84.43% <27.77%> (-0.95%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmcls/utils/distribution.py	10.52% <10.52%> (ø)
mmcls/apis/train.py	15.38% <33.33%> (+0.44%)	⬆️
mmcls/utils/device.py	36.36% <36.36%> (ø)
mmcls/__init__.py	89.65% <100.00%> (ø)
mmcls/utils/__init__.py	100.00% <100.00%> (ø)
mmcls/models/backbones/convnext.py	60.00% <0.00%> (-34.79%)	⬇️
mmcls/utils/setup_env.py	72.72% <0.00%> (-22.73%)	⬇️
mmcls/models/utils/helpers.py	79.16% <0.00%> (-20.84%)	⬇️
mmcls/datasets/custom.py	88.52% <0.00%> (-11.48%)	⬇️
mmcls/models/backbones/convmixer.py	88.52% <0.00%> (-9.84%)	⬇️
... and 4 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 11df205...9a8c7f3. Read the comment docs.