Update pre-commit hooks and versions for clang-format, black, and isort

[NripeshN]

No description provided.

[Copilot]

Pull request overview

This PR updates pre-commit hook versions for clang-format, black, and isort, which triggered automatic formatting changes across the codebase. The removed pre-commit-hooks repository contained only a check-yaml hook and commented-out hooks.

Changes:

• Updated clang-format version from v19.1.7 to v21.1.8
• Updated black version from 25.1.0 to 26.1.0
• Updated isort version from 6.0.0 to 7.0.0
• Removed the pre-commit-hooks repository section (which only had check-yaml enabled)
• Applied automatic formatting changes throughout C++ and Metal shader code

Reviewed changes

Copilot reviewed 28 out of 37 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
.pre-commit-config.yaml	Updated tool versions and removed pre-commit-hooks repository
tests/linalg_tests.cpp	Reformatted multi-line function calls for better alignment
python/src/ops.cpp	Reformatted nanobind signature strings across multiple lines
mlx/distributed/ring/ring.cpp	Reformatted nested function calls and lambda expressions
mlx/distributed/jaccl/jaccl.cpp	Reformatted TCPSocket::connect call
mlx/backend/metal/matmul.cpp	Reformatted struct initializers to be more compact
mlx/backend/metal/kernels/steel/gemm/mma.h	Reformatted function parameter alignment
mlx/backend/metal/kernels/steel/gemm/kernels/*.h	Added spacing in template expressions (WM*WN → WM * WN)
mlx/backend/metal/kernels/gemv_masked.h	Added spacing in template expressions
mlx/backend/metal/kernels/gemv.metal	Added spacing in template expressions
mlx/backend/metal/kernels/conv.metal	Reformatted macro definitions for better readability
mlx/backend/cuda/unary/unary.cuh	Reformatted error message formatting
mlx/backend/cuda/sort.cu	Reformatted CUB library function calls
mlx/backend/cuda/scan.cu	Reformatted error message formatting
mlx/backend/cuda/lru_cache.h	Reformatted error message formatting
mlx/backend/cuda/jit_module.cpp	Reformatted error message formatting
mlx/backend/cuda/indexing.cpp	Reformatted fmt::format calls
mlx/backend/cuda/gemms/cublas_gemm.cpp	Reformatted error message formatting
mlx/backend/cuda/event.cu	Reformatted constructor initializer list
mlx/backend/cuda/device.cpp	Reformatted error message formatting
mlx/backend/cuda/cutlass_utils.cuh	Reformatted error message formatting
mlx/backend/cuda/cudnn_utils.h	Reformatted error message and tensor creation
mlx/backend/cuda/cublas_utils.h	Reformatted error message formatting
mlx/backend/cuda/compiled.cpp	Reformatted fmt::format calls
mlx/backend/cuda/binary_two.cu	Reformatted error message formatting
mlx/backend/cuda/binary/binary.cuh	Reformatted error message formatting
mlx/backend/cpu/jit_compiler.cpp	Reformatted JitCompiler::exec calls
mlx/backend/cpu/compiled.cpp	Reformatted JitCompiler calls and error messages
mlx/array.h	Reformatted constructor initializer list
mlx/array.cpp	Reformatted constructor initializer lists

---

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

[awni]

LGTM

[awni]

Mind rebasing and then we can run the tests?

[awni]

I might recommend starting from a fresh branch and then:

1. Applying the version bump to the linters
2. Run the linters again

Rather than trying to patch the format changes.

[NripeshN]

I might recommend starting from a fresh branch and then:

1. Applying the version bump to the linters
2. Run the linters again

Rather than trying to patch the format changes.

Ah I usually take the lazier approach, copy the conflicting files from main and paste it into my branch manually and run pre-commit😅.