fix custom module for RNNModel and add tests

[dennisbader]

Fixes #1523, fixes #2082.

Summary

• adds CustomRNNModule and CustomBlockRNNModule for defining custom rnn models.
• To create a new module, subclass from :class:CustomBlockRNNModule and:
  • Define the architecture in the module constructor (__init__())
  • Add the forward() method and define the logic of your module's forward pass
  • Use the custom module class when creating a new Block/RNNModel with parameter model.
• fixes error when passing a custom rnn module to RNNModel. The model expected an instantiated module object, but then attempted to instantiate the module from the object when creating an RNNmodel (resulting in calling the forward() pass instead of the module constructor).

[codecov-commenter]

Codecov Report

Attention: 3 lines in your changes are missing coverage. Please review.

Comparison is base (7cfdf62) 93.91% compared to head (facbc4e) 93.91%.
Report is 2 commits behind head on master.

Files	Patch %	Lines
darts/models/forecasting/block_rnn_model.py	96.42%	1 Missing ⚠️
darts/models/forecasting/rnn_model.py	96.96%	1 Missing ⚠️
...arts/models/forecasting/torch_forecasting_model.py	50.00%	1 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2088   +/-   ##
=======================================
  Coverage   93.91%   93.91%           
=======================================
  Files         135      135           
  Lines       13299    13319   +20     
=======================================
+ Hits        12490    12509   +19     
- Misses        809      810    +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[madtoinou]

LGTM, thank you @dennisbader for taking care of this!