update swav to override optimizer_step with optimizer.step(closure=op…

[ananyahjha93]

What does this PR do?

fixes swav with optimizer.step(closure=optimizer_closure). Otherwise training_step was being skipped.

blocked by/tests failing due to: Lightning-AI/pytorch-lightning#4455
needs lightning to update to this PR.

[codecov]

Codecov Report

Merging #323 into master will decrease coverage by 0.12%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #323      +/-   ##
==========================================
- Coverage   82.22%   82.09%   -0.13%     
==========================================
  Files          98       98              
  Lines        5491     5491              
==========================================
- Hits         4515     4508       -7     
- Misses        976      983       +7     

Flag	Coverage Δ
cpu	24.25% <66.66%> (+0.05%)	⬆️
pytest	24.25% <66.66%> (+0.05%)	⬆️
unittests	81.47% <100.00%> (-0.13%)	⬇️

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

Impacted Files	Coverage Δ
...l_bolts/models/self_supervised/swav/swav_module.py	74.34% <100.00%> (+0.43%)	⬆️
pl_bolts/optimizers/lars_scheduling.py	78.72% <0.00%> (-17.03%)	⬇️

---

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 e6b1087...4c5ab22. Read the comment docs.