fix(array): don't panic on unsupported arrow types

[miniex]

Summary

DType::from_arrow hit the unimplemented! fallback arm for unsupported Arrow types (Duration, Interval, FixedSizeBinary), which surfaced in Python as a PanicException instead of a clean error. I added a fallible TryFromArrowType conversion that returns a VortexError on that arm, kept the infallible FromArrowType as a thin delegate for existing trusted callers, and routed the ArrowSession leaf path and the Python entry points through it so unsupported types raise a clean ValueError.

The error keeps the not supported marker so it lands in the fuzzer's existing unsupported-type skip path; the Duration/FixedSizeBinary generation exclusions can be lifted once the fuzzer is on develop.

Closes: #8346

Testing

Added an rstest in vortex-array/src/dtype/arrow.rs checking Duration/Interval/FixedSizeBinary error cleanly as a leaf and nested in a field/schema, plus a parametrized regression test in vortex-python/test/test_array.py for the issue's repro; both of the issue's Python snippets now raise ValueError instead of panicking. cargo nextest run -p vortex-array (3060 tests) and uv run --all-packages pytest vortex-python/test/test_array.py pass; fmt --all, clippy --all-targets --all-features, cargo doc, basedpyright and ruff are clean.

---

I'm Korean, so sorry if any wording reads a little awkward.

[AdamGS]

This seems good, just running CI to see if everything else is good.

[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.

⚡ 2 improved benchmarks
❌ 2 regressed benchmarks
✅ 1581 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_canonical_into[(1000, 10)]	154.8 µs	190.9 µs	-18.93%
❌	Simulation	slice_empty_vortex	310 ns	368.3 ns	-15.84%
⚡	Simulation	chunked_varbinview_into_canonical[(1000, 10)]	204.4 µs	168.7 µs	+21.17%
⚡	Simulation	bitwise_not_vortex_buffer_mut[128]	273.6 ns	244.4 ns	+11.93%

Tip

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

---

_{Comparing miniex:fix/arrow-unsupported-dtype (136c0a6) with develop (4668b6e)}

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. ↩

[robert3005]

Shouldn't this be migrated to the session arrow conversion? it's a bigger change