Allow configuration of nn ensemble learning rate

[mo-fu]

When training nn-ensemble with optimizers that do not automatically adjust the learning rate it is helpful to configure it manually. Even with automatically adjusting optimizers, it may sometimes be beneficial to change the learning rate.
This pr allow to set the learning rate in the nn-ensemble project definition.
As different optimizers have different default learning rates, it is not possible to just add a default parameter. Instead the presence of a lr key in the config map is checked.

[sonarcloud]

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
0.0% Duplication

[codecov]

Codecov Report

Merging #475 (b8e556c) into master (5bf40f8) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #475   +/-   ##
=======================================
  Coverage   99.43%   99.44%           
=======================================
  Files          67       67           
  Lines        4807     4838   +31     
=======================================
+ Hits         4780     4811   +31     
  Misses         27       27           

Impacted Files	Coverage Δ
annif/backend/nn_ensemble.py	100.00% <100.00%> (ø)
tests/test_backend_nn_ensemble.py	100.00% <100.00%> (ø)
annif/suggestion.py	100.00% <0.00%> (ø)
tests/test_suggestion.py	100.00% <0.00%> (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 5bf40f8...b8e556c. Read the comment docs.

[osma]

Thanks @mo-fu !