fix(spark): include the offending path when schema inference finds no Vortex files

[jackylee-ch]

Rationale for this change

When VortexDataSourceV2.inferSchema is pointed at a directory that contains no
.vortex files, it currently throws
RuntimeException("UNABLE_TO_INFER_SCHEMA format: vortex"). The message does not say
which path failed, so with multi-path reads (or a typo'd location) users have nothing
to go on. This is the same kind of diagnosability gap as #8759 (unsupported ColumnVector
type error message).

What changes are included in this PR?

• inferSchema now throws IllegalArgumentException with a message that names the
  offending path and suggests the likely fixes:
  Unable to infer schema for vortex: no .vortex files found under path <path>. Check that the path is correct and contains at least one Vortex file, or provide an explicit schema.
• The @throws javadoc on inferSchema is updated to match.
• A new VortexDataSourceInferSchemaTest reads an empty temp directory through the real
  spark.read().format("vortex").load(...) path and asserts the failure message names
  the path (walking the cause chain, since Spark may wrap data source exceptions).

Passing locally on both matrix legs (:vortex-spark_2.12:test and
:vortex-spark_2.13:test with --tests 'dev.vortex.spark.VortexDataSourceInferSchemaTest').

What APIs are changed? Are there any user-facing changes?

The exception type thrown by inferSchema for the no-files case changes from a bare
RuntimeException to IllegalArgumentException (still a RuntimeException subtype),
and the message text changes as described. No other behavior changes.

[codspeed-hq]

Merging this PR will improve performance by 11.17%

⚡ 1 improved benchmark
✅ 1669 untouched benchmarks
⏩ 44 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	true_count_vortex_buffer[128]	580.6 ns	522.2 ns	+11.17%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing jackylee-ch:spark-infer-schema-error-msg (94dd273) with develop (b7b01d3)}

Footnotes

1. 44 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. ↩