Add SIMD support via xsimd library

[zfergus]

Description

This pull request introduces cross-platform SIMD (Single Instruction, Multiple Data) support using the xsimd library, replacing the previous Apple-specific SIMD implementation in the LBVH code. It also adds build system enhancements to detect and configure SIMD support automatically. The main changes are grouped below:

Build System Enhancements

• Update the IPC_TOOLKIT_WITH_SIMD to be enabled by default, and improve CMake logic to detect SIMD capabilities and configure the build accordingly.
  • If SIMD is not found, the option is turned off with a warning.
• Integrated the xsimd library as a dependency using a new CMake recipe (cmake/recipes/xsimd.cmake), and updated the build to link against xsimd and disable Eigen's internal vectorization to avoid conflicts.

Codebase Modernization for SIMD:

• Refactored SIMD code in src/ipc/broad_phase/lbvh.cpp to use xsimd for cross-platform SIMD support, replacing Apple-specific simd intrinsics. This includes updating batch types, boolean operations, and utility functions to use xsimd APIs.
• Updated the configuration header template to include the new SIMD option macro. (src/ipc/config.hpp.in)

These changes make SIMD acceleration available on all supported platforms, not just Apple, and improve maintainability by standardizing on a widely-used SIMD abstraction library.

Type of change

Please delete options that are not relevant.

• Enhancement (non-breaking change which improves existing functionality)
• This change requires a documentation update

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.35%. Comparing base (a9ece21) to head (259c434).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #207   +/-   ##
=======================================
  Coverage   97.34%   97.35%           
=======================================
  Files         161      161           
  Lines       24664    24697   +33     
  Branches      877      883    +6     
=======================================
+ Hits        24010    24044   +34     
+ Misses        654      653    -1     

Flag	Coverage Δ
unittests	97.35% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Pull request overview

This pull request modernizes the SIMD implementation in IPC Toolkit by replacing Apple-specific SIMD intrinsics with the cross-platform xsimd library, enabling SIMD acceleration on all supported platforms (not just macOS). The changes also make SIMD enabled by default with automatic fallback when not available.

Changes:

• Integrated xsimd library (v14.0.0) as a dependency for cross-platform SIMD support
• Refactored LBVH SIMD code to use xsimd APIs with architecture-dependent batch sizes
• Updated build system to enable SIMD by default with automatic capability detection and graceful degradation

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
cmake/recipes/xsimd.cmake	Adds xsimd library integration via CPM
CMakeLists.txt	Enables SIMD by default, adds detection logic, links xsimd, and disables Eigen vectorization
CMakePresets.json	Removes dedicated SIMD preset (now default) and SIMD-OFF flag from python preset
IPCToolkitOptions.cmake.sample	Reorders SIMD option to reflect its new default-enabled status
src/ipc/config.hpp.in	Adds IPC_TOOLKIT_WITH_SIMD configuration macro
src/ipc/utils/eigen_ext.hpp	Adds workaround for Eigen abs() issue when vectorization is disabled
src/ipc/broad_phase/lbvh.cpp	Refactors SIMD code from Apple-specific to xsimd with dynamic batch sizes

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The CPMAddPackage("gh:xtensor-stack/xsimd#14.0.0") call introduces a supply-chain risk by fetching and building third-party code from GitHub pinned only to a mutable tag without any additional integrity verification. If the xtensor-stack/xsimd repository or its tags are compromised, CI/builds that have access to secrets could execute attacker-controlled code. To mitigate this, pin the dependency to an immutable commit SHA (and, if supported by CPM, enable checksum/signature verification) rather than relying solely on a version tag.