Add prefix arg to metric collection

[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?

Fixes #64
Adds a prefix argument to the metric collection for easy initialization of the same metric collection multiple times with adjusted keys

mc = MetricCollection([Accuracy, Precision, Recall], prefix='train_')
out = mc(...)
# keywords of output dict will be "train_Accuracy", "train_Precision", "train_Recall"

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 #70 (4c7d878) into master (75570f6) will decrease coverage by 0.03%.
The diff coverage is 95.00%.

@@            Coverage Diff             @@
##           master      #70      +/-   ##
==========================================
- Coverage   97.11%   97.07%   -0.04%     
==========================================
  Files          59       59              
  Lines        1938     1949      +11     
==========================================
+ Hits         1882     1892      +10     
- Misses         56       57       +1     

Flag	Coverage Δ
Linux	78.54% <80.00%> (-0.04%)	⬇️
Windows	78.54% <80.00%> (-0.04%)	⬇️
cpu	97.07% <95.00%> (-0.04%)	⬇️
macOS	97.07% <95.00%> (-0.04%)	⬇️
pytest	97.07% <95.00%> (-0.04%)	⬇️
python3.6	97.06% <95.00%> (-0.04%)	⬇️
python3.8	97.07% <95.00%> (-0.04%)	⬇️
python3.9	97.07% <95.00%> (-0.04%)	⬇️
torch1.3.1	95.31% <95.00%> (-0.03%)	⬇️
torch1.4.0	95.48% <95.00%> (-0.03%)	⬇️
torch1.8.0	97.07% <95.00%> (-0.04%)	⬇️

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

Impacted Files	Coverage Δ
torchmetrics/collections.py	97.91% <94.73%> (-2.09%)	⬇️
torchmetrics/__init__.py	100.00% <100.00%> (ø)

---

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 75570f6...4c7d878. Read the comment docs.

[carmocca]

LGTM. Can you also add return type annotations?