test: add dual-environment validation tests for pandas-gbq delegation

[shuoweil]

Adds end-to-end unit test validation for ReadRowsPage.to_arrow() delegation to pandas-gbq and fallback behavior when pandas-gbq is uninstalled.

Changes

• Added packages/google-cloud-bigquery-storage/tests/unit/test_reader_pandas_gbq.py
• Tests real protobuf decoding and Arrow RecordBatch assembly across multiple BigQuery data types (ints, floats, strings, decimals, dates, UTC timestamps).
• Tests empty batch response handling.
• Tests multi-page stream concatenation.
• Tests safe fallback to legacy in-tree parser when pandas-gbq is absent.

Fixes #< 540939659 > 🦕

[gemini-code-assist]

Code Review

This pull request introduces a new test suite test_reader_pandas_gbq.py to verify Arrow serialization and deserialization behaviors, including delegation to pandas_gbq and fallback paths when it is not installed. The review feedback points out that importing pandas_gbq.arrow at the top level will cause test discovery to fail in environments where pandas-gbq is missing. It suggests dynamically importing the optional dependency using pytest.importorskip within the relevant tests to ensure robust test execution and proper path isolation.

[gemini-code-assist]

Importing pandas_gbq.arrow at the top level will cause the entire test module to fail to load in environments where pandas-gbq is not installed. Since this file contains tests specifically designed to verify the fallback behavior when pandas-gbq is not installed (e.g., test_read_rows_page_to_arrow_falls_back_when_pandas_gbq_uninstalled), we should avoid top-level imports of optional dependencies. Instead, import them dynamically within the specific tests that require them.

[shuoweil]

fixed