Use CCCL overflow utils instead of custom overflow logic

[PointKernel]

Description

Uncovered this cleanup opportunity while working on #22465.

This PR replaces cudf's custom overflow handling logic with the overflow utilities provided by CCCL. Note that although these utilities are under the cuda:: namespace, they are available on both host and device.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR refactors overflow detection across cuDF by replacing manual numeric_limits-based comparisons with CUDA's native overflow intrinsics. Five files now include <cuda/numeric> and use cuda::add_overflow, cuda::sub_overflow, cuda::div_overflow, and cuda::mul_overflow to detect arithmetic overflow with their .overflow result member.

Changes

Overflow Detection Refactoring

Layer / File(s)	Summary
Fixed-point integer overflow helpers cpp/include/cudf/fixed_point/fixed_point.hpp	Copyright year updated to 2026. Added <cuda/numeric> include. Four overflow helper functions (addition_overflow, subtraction_overflow, division_overflow, multiplication_overflow) refactored to use CUDA intrinsics' .overflow member instead of manual numeric_limits range checks.
Device aggregators SUM_WITH_OVERFLOW overflow detection cpp/include/cudf/detail/aggregation/device_aggregators.cuh	Header include changed from cuda/std/limits to cuda/numeric. Removed type_max/type_min constants from update_target_element<Source, aggregation::SUM_WITH_OVERFLOW> specialization. Overflow detection after atomic add now uses cuda::add_overflow<DeviceType>(...).overflow.
Bitmask null_mask cumulative overflow check cpp/src/bitmask/null_mask.cu	Added <cuda/numeric> include. In set_null_masks, cumulative word overflow detection during pinned-vector tabulation replaced with cuda::add_overflow-based check.
ORC decompression output size overflow check cpp/src/io/orc/orc.cpp	Added <cuda/numeric> include. In orc_decompressor::decompress_blocks, worst-case output size computation now uses cuda::add_overflow<size_t> instead of manual numeric_limits-based bounds check.
Parquet byte bounds filtering overflow check cpp/src/io/parquet/reader_impl_helpers.cpp	Added <cuda/numeric> include. In aggregate_reader_metadata::apply_byte_bounds_filter, upper-bound overflow check for bytes_to_skip + bytes_to_read changed to cuda::add_overflow<size_t>(...) instead of comparing against numeric_limits::max().

---

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the main change: replacing custom overflow handling with CCCL utilities.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description check	✅ Passed	The pull request description clearly describes the changeset: replacing custom overflow handling with CCCL overflow utilities across multiple files.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[vyasr]

Signpost for the next reviewer, this line is the meaningful change in this PR. The rest is just clang-format.

[mhaseeb123]

Nice work. Also TIL cuda::xx_overflow<>().overflow cccl utils

[PointKernel]

/merge