Remove use of deepcopy

[SkafteNicki]

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure to update the docs?
• Did you write any new necessary tests?

What does this PR do?

After profiling some metrics, I saw that that reset is spending a lot of time calling deepcopy. As the recommended method for cloning a tensor is .detach().clone() and it is a bit faster:

master:
tensor state (cpu): 0.8742028713226319+-0.03886386210968593
tensor state (gpu): 2.5115918636322023+-0.21529070194941985
list state (cpu): 2.109953451156616+-0.07316269549459964
list state (gpu): 1.809204387664795+-1.0394736704878353

PR:
tensor state (cpu):0.8499699592590332+-0.01647528474876866
tensor state (gpu):2.2886361598968508+-0.08935389958063863
list state (cpu): 2.0209419250488283+-0.05932843043190113
list state (gpu): 1.547454833984375+-0.407560693521206

calculate for 5 repetitions over 1000 forward calculations (reset is called once for every forward call which is why such a small optimization may matter in the long run).

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #163 (4ac6a02) into master (a17ccb2) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##           master     #163     +/-   ##
=========================================
  Coverage   95.99%   95.99%             
=========================================
  Files         168       84     -84     
  Lines        5144     2572   -2572     
=========================================
- Hits         4938     2469   -2469     
+ Misses        206      103    -103     

Flag	Coverage Δ
Linux	78.73% <100.00%> (ø)
Windows	78.73% <100.00%> (ø)
cpu	95.99% <100.00%> (ø)
gpu	?
macOS	95.99% <100.00%> (ø)
pytest	95.99% <100.00%> (ø)
python3.6	95.98% <100.00%> (ø)
python3.8	95.99% <100.00%> (ø)
python3.9	95.99% <100.00%> (ø)
torch1.3.1	94.85% <100.00%> (ø)
torch1.4.0	94.98% <100.00%> (ø)
torch1.8.1	95.99% <100.00%> (ø)

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

Impacted Files	Coverage Δ
torchmetrics/metric.py	95.07% <100.00%> (ø)
...etrics/functional/regression/explained_variance.py
__w/1/s/torchmetrics/regression/ssim.py
...chmetrics/functional/classification/stat_scores.py
...s/torchmetrics/classification/average_precision.py
...ics/functional/classification/average_precision.py
...1/s/torchmetrics/functional/retrieval/precision.py
__w/1/s/torchmetrics/__init__.py
.../s/torchmetrics/classification/hamming_distance.py
__w/1/s/torchmetrics/functional/image_gradients.py
... and 75 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 a17ccb2...4ac6a02. Read the comment docs.

[tchaton]

LGTM !