add support for more lr scheduler config parameters to torch models

[dennisbader]

Summary

Adds support for more lr scheduler configuration parameters to TorchForecastingModels:

• "interval": "epoch" (The unit of the scheduler's step size, could also be 'step'.
  'epoch' updates the scheduler on epoch end whereas 'step'
  updates it after a optimizer update)
• "frequency": 1 (How many epochs/steps should pass between calls to
  scheduler.step(). 1 corresponds to updating the learning
  rate after every epoch/step.)
• "monitor": "val_loss" (Metric to to monitor for schedulers like ReduceLROnPlateau)
• "strict": True (If set to True, will enforce that the value specified 'monitor'
  is available when the scheduler is updated, thus stopping
  training if not found. If set to False, it will only produce a warning)
• "name": None (If using the LearningRateMonitor callback to monitor the
  learning rate progress, this keyword can be used to specify
  a custom logged name)

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (8cb04f6) 93.88% compared to head (45f062a) 93.87%.
Report is 2 commits behind head on master.

❗ Current head 45f062a differs from pull request most recent head a810eb5. Consider uploading reports for the commit a810eb5 to get more accurate results

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

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2218      +/-   ##
==========================================
- Coverage   93.88%   93.87%   -0.02%     
==========================================
  Files         135      135              
  Lines       13425    13415      -10     
==========================================
- Hits        12604    12593      -11     
- Misses        821      822       +1     

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