[Data] Fix error message and off-by-one bounds check in operators

[jiangxt2]

Why are these changes needed?

Three minor bugs in Ray Data operators:

1. _concatenate_chunked_arrays prints type: None in its assertion error message instead of the actual array type. The f-string uses {type_} which is always None at that point in the loop — it is initialized to None and only reassigned after the assertion.

2. UnionOperator._add_input_inner uses <= instead of < in its bounds check, allowing input_index == len(self._input_dependencies). This is semantically incorrect — input_index is 0-based, so valid values are [0, len-1]. The <= allows an out-of-bounds index to pass the assertion.

3. ZipOperator._add_input_inner has the same off-by-one bounds check.

Both off-by-one bugs are corrected to match the pattern used by MixOperator._add_input_inner (line 95), which correctly uses strict <.

What changes are made?

• python/ray/data/_internal/arrow_ops/transform_pyarrow.py: Change {type_} to {arr.type} in assertion error message
• python/ray/data/_internal/execution/operators/union_operator.py: Change <= to < in bounds check
• python/ray/data/_internal/execution/operators/zip_operator.py: Change <= to < in bounds check
• python/ray/data/tests/unit/test_fix_operator_bounds_check.py: Add 3 tests covering the fixes

Related issues

N/A

Checks

• I've signed off every commit (using git commit -s)
• I've run pre-commit hooks (ruff, black, semgrep, pydoclint) to lint the changes
• I've added tests for the new behavior

[gemini-code-assist]

Code Review

This pull request addresses three bugs: it corrects an assertion error message in _concatenate_chunked_arrays to display the actual tensor type instead of None, and fixes off-by-one bounds checking errors in both UnionOperator and ZipOperator input addition methods. It also introduces corresponding unit tests. The feedback suggests refactoring the _concatenate_chunked_arrays test to use the more idiomatic pytest.raises context manager instead of a try-except block with pytest.fail.

[github-actions]

This pull request has been automatically marked as stale because it has not had
any activity for 14 days. It will be closed in another 14 days if no further activity occurs.
Thank you for your contributions.

You can always ask for help on our discussion forum or Ray's public slack channel.

If you'd like to keep this open, just leave any comment, and the stale label will be removed.