[DOC] Migrate _rnn.py docstrings from Google style to NumPyDoc style

[Abhishek242004]

Reference Issues/PRs

Fixes #2066

What does this implement/fix? Explain your changes.

This PR migrates the docstrings in pytorch_forecasting/models/.../_rnn.py from Google style to NumPy docstring style to align with the project's documentation standards and the ongoing migration described in #2066.

The following updates were made:

• Converted Google-style sections such as Args: and Returns: to NumPy-style sections like Parameters, Returns, and Attributes.
• Improved formatting and structure to comply with NumPyDoc conventions used across the repository.

These changes are purely documentation improvements and do not affect the functionality of the code.

What should a reviewer concentrate their feedback on?

• Correctness and consistency of the NumPy docstring formatting
• Accuracy of parameter and return descriptions
• Correctness of documented tensor shapes in the forward methods

Did you add any tests for the change?

No. This is a documentation-only change.

Any other comments?

• Verified that all pre-commit hooks pass locally.

PR checklist

• The PR title starts with either [ENH], [MNT], [DOC], or [BUG].
• Added/modified tests
• Used pre-commit hooks when committing to ensure that code is compliant with hooks. Install hooks with pre-commit install.
  To run hooks independent of commit, execute pre-commit run --all-files
  -->

[phoeenniixx]

The code-quality is failing, please use pre-commit before pushing
https://www.sktime.net/en/stable/developer_guide/coding_standards.html#setting-up-local-code-quality-checks