build(deps): bump the cargo group in /rust with 2 updates

[dependabot]

Updates the requirements on onnx-runtime-ir and bindgen to permit the latest version.
Updates onnx-runtime-ir to 0.1.0-dev.6

Release notes

Sourced from onnx-runtime-ir's releases.

nxrt-ep-v0.1.0-dev.6

Development release of the ORT plugin-EP packages, tagged from exact candidate 49590f7c22da2ee554aa131e394e8c5f50ef7961.

Publishes nxrt-ep-cpu==0.1.0.dev6 for Linux x86_64, Windows AMD64, Windows ARM64, and macOS arm64, and experimental nxrt-ep-cuda==0.1.0.dev6 for Linux x86_64. CUDA runtime libraries are dynamically loaded rather than bundled.

Release wheels are externally available on PyPI; corresponding build artifacts remain attached to Actions run 34449874469.

Commits

• See full diff in compare view

Updates bindgen to 0.73.2

Changelog

Sourced from bindgen's changelog.

0.73.2 (2026-09-08)

Added

Changed

Removed

Fixed

• Fixed various regressions with over/underaligned types with respect to 0.72 (#3449,#3453,#3458).
• Fixed a codegen regression with bitfields with enum types in some circumstances (#3455).

Security

0.73.1 (2026-09-05)

Added

Changed

• Widened acceptable prettyplease / syn dependency versions (#3447).

Removed

Fixed

Security

0.73.0 (2026-09-05)

Added

• Add Debug, Clone and Eq traits to ParseCallbacks info types (#3233).
• Include the source location in discovery ParseCallbacks (#3244).
• Report constants and variables in ParseCallbacks::new_item_found (#3249).
• Allow adding custom attributes to extern function blocks (#3257).
• Support new-type-alias constants (#3288, #3390).
• Support nested flexible array members (#3306).
• Add support for custom per-field attributes (#3307).
• Add const-generic bitfield accessors for compile-time optimization (#3325, #3374).
• Allow renaming struct and union fields (#3358).
• Add a libcpp feature to link libclang against libc++ (#3375).
• Prebuilt binaries for aarch64 Linux (#3217).

Changed

• Guarantee opaque type layout across all architectures (#3280).
• Use NonNull for C++ references (#3292).
• Bindings::write now takes impl Write instead of Box<dyn Write> (#3317, #3324).
• Optimize bitfield unit get/set with byte-wise operations (#3325).
• Emit const __BindgenUnionField accessors and comply with Rust 2024 unsafe_op_in_unsafe_fn (#3337).
• Generate a doc comment pointing at the aliased type when a type alias is hidden (#3122).
• dyngen: convert the library path to OsStr before opening it (#3315).
• Return a NoHeadersProvided error instead of generating empty bindings when no input headers are provided (#3391).
• Bump the MSRV to 1.71 (#3401).
• Bump syn to v3 (#3401) and shlex to v2 (#3380).

Removed

• Removed support for generating code for rustc versions < 1.51, along with

... (truncated)

Commits

• 73c69d6 chore: Release
• 77cbc72 codegen: Pad bitfields to field offset if applicable.
• dc61531 ty: More struct layout fixes for overaligned integers / aliases.
• 4348523 codegen: Revert some bitfield codegen changes and silence unnecessary_transmu...
• efb6e0c ty: Use underlying layout of typedefs if available.
• 18fc3ac codegen: Simplify bitfield merging code.
• 910679f codegen: Remove hack that is no longer needed now that we have repr(align) ev...
• 38ebc48 codegen: Don't force padding to be aligned.
• 64f0939 codegen: Don't force padding for over-aligned fields.
• cf8faab codegen: Simplify some conditions about repr(align).
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[github-actions]

🏎️ MLX EP op benchmark

Median session.run (MLX EP), base vs PR — 3 regression(s) 🔴 · 8 improvement(s) 🟢 (threshold ±10%).

Case	base ms	PR ms	Δ%	note
matmul_fp32_256x512x512	0.962	1.250	+30.0%	🔴 regression
matmul_fp16_256x512x512	1.024	1.306	+27.5%	🔴 regression
matmulnbits_decode_M1_K2048_N2048	0.885	1.051	+18.7%	🔴 regression
matmul_fp16_gemv_1x2048x2048	2.257	2.419	+7.2%
rmsnorm_512x1024	0.924	0.936	+1.4%
diffusion_resnet_block_320ch_64x64	22.415	22.481	+0.3%
simplified_ln_default_domain_512x1024	0.854	0.843	-1.2%
matmulnbits_prefill_M256_K1024_N1024	1.538	1.427	-7.2%
diffusion_upblock_resize_concat_320ch_32x32	23.206	18.213	-21.5%	🟢 improvement
layernorm_fp32_512x1024	1.218	0.903	-25.8%	🟢 improvement
conv2d_fp32_32to64_64x64	1.437	1.025	-28.6%	🟢 improvement
gqa_prefill_s256_h16kv4d64	2.994	1.897	-36.7%	🟢 improvement
softmax_fp32_1x16x256x256	2.037	1.248	-38.7%	🟢 improvement
gelu_fp32_512x1024	1.338	0.816	-39.0%	🟢 improvement
add_broadcast_fp16_512x1024	1.204	0.697	-42.1%	🟢 improvement
gqa_decode_s1_past256_h16kv4d64	2.994	1.311	-56.2%	🟢 improvement

_{base base (main) · PR PR · 60 iters (median) · ORT 1.29.0 · lower ms is better. Timings on a shared runner are noisy; only |Δ%| > 10% is flagged.}