chore: Python CUDA bridge: CI and buffer handoff ABI

[0ax1]

CI coverage for the Python CUDA bridge through the vortex-data[cuda] optional-extra path.

Extend the private metadata bridge to carry host buffer-export capsules instead of only a buffer count. The base Python package exports repr(C) VortexBufferExport descriptors, and vortex-python-cuda imports them into local BufferHandles before deserializing arrays through its own VortexSession.

Tests now cover primitive, nullable, bool, and struct arrays across the bridge, plus the existing CUDA Arrow Device smoke path.

[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
❌ 2 regressed benchmarks
✅ 1589 untouched benchmarks
⏩ 4 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	chunked_varbinview_into_canonical[(1000, 10)]	169 µs	205.6 µs	-17.79%
❌	WallTime	cuda/bitpacked_u8/unpack/3bw[100M]	299.5 µs	350.9 µs	-14.66%
⚡	Simulation	chunked_varbinview_canonical_into[(100, 100)]	259.4 µs	224.3 µs	+15.65%
⚡	Simulation	chunked_varbinview_into_canonical[(100, 100)]	306.7 µs	271.4 µs	+13%
⚡	Simulation	bitwise_not_vortex_buffer_mut[128]	273.6 ns	244.4 ns	+11.93%
⚡	Simulation	eq_i64_constant	319.3 µs	289.4 µs	+10.32%

Tip

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

---

_{Comparing ad/pycudf3 (f2bb115) with develop (2a15a9f)}

Footnotes

1. 4 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩