docs: add connector libraries overview table to package README

[shuoweil]

The google-cloud-bigquery reference page serves as the package README. Adding a connector libraries table guides users seeking DataFrame operations to high-level BigQuery libraries.

Changes

• Added a Related Libraries section to packages/google-cloud-bigquery/README.rst.
• Linked and described key ecosystem connectors:
  • BigQuery DataFrames (bigframes): Scalable pandas-like and scikit-learn-like API powered by BigQuery.
  • pandas-gbq: Lightweight integration for read_gbq() and to_gbq().

Fixes #<522853190> 🦕

[gemini-code-assist]

Code Review

This pull request deprecates retrieving Arrow record batches directly via google-cloud-bigquery-storage, delegating this functionality to a new arrow module in pandas-gbq. It also adds related documentation and updates tests. The review feedback identifies two important issues in the new pandas_gbq/arrow.py file: a potential AttributeError when accessing arrow_record_batch if it is None, and an invalid attribute access on pyarrow.ipc.Message, which does not possess a schema attribute.

[gemini-code-assist]

A pyarrow.ipc.Message object does not have a schema attribute, so getattr(msg, "schema", ...) will always fall back to the default value. Since we cannot retrieve the schema from the message itself in this context, we should simplify this to use arrow_schema directly if available, or fall back to an empty schema.

        batch_schema = arrow_schema if arrow_schema is not None else pa.schema([])

[shuoweil]

Fixed.