feat: add cuDF C FFI

[0ax1]

No description provided.

[codspeed-hq]

Merging this PR will not alter performance

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚡ 4 improved benchmarks
❌ 5 regressed benchmarks
✅ 1510 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	chunked_varbinview_canonical_into[(1000, 10)]	163.1 µs	198.6 µs	-17.87%
❌	Simulation	chunked_varbinview_into_canonical[(1000, 10)]	177.1 µs	213.5 µs	-17.07%
❌	Simulation	chunked_varbinview_opt_canonical_into[(1000, 10)]	191.4 µs	226.9 µs	-15.66%
❌	Simulation	chunked_varbinview_opt_into_canonical[(1000, 10)]	205.7 µs	242 µs	-14.99%
❌	Simulation	bitwise_not_vortex_buffer_mut[128]	216.9 ns	246.1 ns	-11.85%
⚡	Simulation	chunked_varbinview_canonical_into[(100, 100)]	311.3 µs	276.1 µs	+12.75%
⚡	Simulation	encode_varbin[(1000, 32)]	168.3 µs	150.4 µs	+11.92%
⚡	Simulation	encode_varbin[(1000, 8)]	162.2 µs	145 µs	+11.85%
⚡	Simulation	chunked_varbinview_into_canonical[(100, 100)]	363.2 µs	328.1 µs	+10.7%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing ad/cudf-c-ffi (363b4ec) with develop (d650cab)}

[myrrc]

LGTM

[robert3005]

I think you need to gate some tests on cuda availability

    thread 'tests::test_export_reports_cuda_initialization_error' (88195) panicked at /home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/cudarc-0.19.7/src/lib.rs:200:5:
    Unable to dynamically load the "cuda" shared library - searched for library names: ["libcuda.so", "libcuda64.so", "libcuda64_12.so", "libcuda64_125.so", "libcuda64_125_0.so", "libcuda64_120_5.so", "libcuda64_10.so", "libcuda64_11.so", "libcuda64_12.so", "libcuda64_120_0.so", "libcuda64_9.so", "libcuda.so.12", "libcuda.so.12", "libcuda.so.11", "libcuda.so.10", "libcuda.so.9", "libcuda.so.1", "libnvcuda.so", "libnvcuda64.so", "libnvcuda64_12.so", "libnvcuda64_125.so", "libnvcuda64_125_0.so", "libnvcuda64_120_5.so", "libnvcuda64_10.so", "libnvcuda64_11.so", "libnvcuda64_12.so", "libnvcuda64_120_0.so", "libnvcuda64_9.so", "libnvcuda.so.12", "libnvcuda.so.12", "libnvcuda.so.11", "libnvcuda.so.10", "libnvcuda.so.9", "libnvcuda.so.1"]. Ensure that `LD_LIBRARY_PATH` has the correct path to the installed library. If the shared library is present on the system under a different name than one of those listed above, please open a GitHub issue.

[0ax1]

I think you need to gate some tests on cuda availability

    thread 'tests::test_export_reports_cuda_initialization_error' (88195) panicked at /home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/cudarc-0.19.7/src/lib.rs:200:5:
    Unable to dynamically load the "cuda" shared library - searched for library names: ["libcuda.so", "libcuda64.so", "libcuda64_12.so", "libcuda64_125.so", "libcuda64_125_0.so", "libcuda64_120_5.so", "libcuda64_10.so", "libcuda64_11.so", "libcuda64_12.so", "libcuda64_120_0.so", "libcuda64_9.so", "libcuda.so.12", "libcuda.so.12", "libcuda.so.11", "libcuda.so.10", "libcuda.so.9", "libcuda.so.1", "libnvcuda.so", "libnvcuda64.so", "libnvcuda64_12.so", "libnvcuda64_125.so", "libnvcuda64_125_0.so", "libnvcuda64_120_5.so", "libnvcuda64_10.so", "libnvcuda64_11.so", "libnvcuda64_12.so", "libnvcuda64_120_0.so", "libnvcuda64_9.so", "libnvcuda.so.12", "libnvcuda.so.12", "libnvcuda.so.11", "libnvcuda.so.10", "libnvcuda.so.9", "libnvcuda.so.1"]. Ensure that `LD_LIBRARY_PATH` has the correct path to the installed library. If the shared library is present on the system under a different name than one of those listed above, please open a GitHub issue.

Yeah, that's part of the reason. Also this needs cleaner separation between vortex-ffi and vortex-cuda which duplicates logic in part.