Loads metric states from state_dict

[hlin09]

The persistent metric states are output in state_dict, but cannot be properly loaded through load_state_dict(). This PR adds handling to loading persisted metric states from state_dict and a unit test.

[ananthsub]

this lgtm - @SkafteNicki what do you think?

[SkafteNicki]

LGTM

[codecov]

Codecov Report

Merging #202 (5578ec9) into master (d432d3d) will decrease coverage by 8.43%.
The diff coverage is 28.57%.

@@            Coverage Diff             @@
##           master     #202      +/-   ##
==========================================
- Coverage   96.65%   88.22%   -8.44%     
==========================================
  Files         180      180              
  Lines        5738     5740       +2     
==========================================
- Hits         5546     5064     -482     
- Misses        192      676     +484     

Flag	Coverage Δ
Linux	79.75% <28.57%> (-0.14%)	⬇️
Windows	79.75% <28.57%> (-0.14%)	⬇️
cpu	79.75% <28.57%> (-16.90%)	⬇️
gpu	96.66% <ø> (+<0.01%)	⬆️
macOS	79.75% <28.57%> (-16.90%)	⬇️
pytest	88.22% <28.57%> (-8.44%)	⬇️
python3.6	?
python3.8	?
python3.9	?
torch1.3.1	?
torch1.4.0	?
torch1.8.1	?

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

Impacted Files	Coverage Δ
torchmetrics/metric.py	55.31% <28.57%> (-39.91%)	⬇️
torchmetrics/utilities/distributed.py	22.85% <0.00%> (-74.29%)	⬇️
torchmetrics/classification/auc.py	47.61% <0.00%> (-52.39%)	⬇️
torchmetrics/functional/classification/auroc.py	46.15% <0.00%> (-40.01%)	⬇️
torchmetrics/functional/regression/psnr.py	60.60% <0.00%> (-36.37%)	⬇️
torchmetrics/functional/classification/accuracy.py	57.57% <0.00%> (-36.37%)	⬇️
...chmetrics/functional/classification/stat_scores.py	66.66% <0.00%> (-33.34%)	⬇️
...rics/functional/classification/confusion_matrix.py	66.66% <0.00%> (-33.34%)	⬇️
torchmetrics/utilities/checks.py	60.54% <0.00%> (-31.90%)	⬇️
... and 44 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 d432d3d...5578ec9. Read the comment docs.