[DOC] Add DeepAR and NBeats usage examples

[tomatotomata]

Summary

This adds focused doctest-style usage examples for the v1 DeepAR and NBeats models and their package containers.

The model examples show the supported from_dataset entry point and link to the complete tutorials or example script. The package examples document the container-to-model resolution used by the estimator suite.

This is intentionally limited to two v1 models, as requested in issue #2377. The remaining models can follow in separate small pull requests.

Validation

• python -m ruff check pytorch_forecasting/models/deepar/_deepar.py pytorch_forecasting/models/deepar/_deepar_pkg.py pytorch_forecasting/models/nbeats/_nbeats.py pytorch_forecasting/models/nbeats/_nbeats_pkg.py passed.
• python -m compileall -q passed for the four changed files.
• Doctest syntax parsing passed for all four new examples.
• git diff --check passed.
• Full import and doctest execution was not available in this Windows checkout because the lightning dependency is not installed.

Closes #2377 for the DeepAR and NBeats slice.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (main@7bcf66c). Learn more about missing BASE report.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2378   +/-   ##
=======================================
  Coverage        ?   87.46%           
=======================================
  Files           ?      175           
  Lines           ?    10165           
  Branches        ?        0           
=======================================
  Hits            ?     8891           
  Misses          ?     1274           
  Partials        ?        0           

Flag	Coverage Δ
cpu	87.46% <ø> (?)
pytest	87.46% <ø> (?)

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

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[phoeenniixx]

Thanks!
Can you please use dataset and show the complete flow of the model training and prediction in the example?

[tomatotomata]

I expanded both model docstrings into runnable end-to-end examples using the built-in autoregressive dataset. Each now creates a TimeSeriesDataSet, derives a validation set, builds the model with from_dataset, trains for one epoch, and calls predict on the validation loader.

The examples are in commit 7a1ab5f. python -m compileall passes for both changed modules. I could not execute the flow here because this checkout does not have the lightning dependency installed. Let me know what you think.

[phoeenniixx]

The code quality is still failing. Also, in place of adding a complete code block, can you add it command-wise? Using >>>? The doctests check >>> and not the code block