GH-46724: [C++][Parquet] OSSFuzz: Prevent from Bad-cast in handling statistics

[mapleFU]

Rationale for this change

See issue

What changes are included in this PR?

Checked before a cast here

Are these changes tested?

no

Are there any user-facing changes?

no

• GitHub Issue: [C++][Parquet] Bad-cast in handling statistics #46724

[github-actions]

⚠️ GitHub issue #46724 has been automatically assigned in GitHub to PR creator.

[pitrou]

Thanks for fixing this. Can you first add the regression file to arrow-testing/data/parquet/fuzzing/?

[pitrou]

@mapleFU Would you like to update this PR?

[mapleFU]

Sorry, I was really busy with apartment hunting for a while. I'll update it today

[mapleFU]

test case: apache/arrow-testing#105

[pitrou]

LGTM, can you just update the parquet-testing submodule?

[pitrou]

Sorry, I was really busy with apartment hunting for a while. I'll update it today

No problem (and I hope the apartment hunting went well)!

[mapleFU]

@pitrou updated

[pitrou]

+1, thanks a lot for fixing this @mapleFU

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 3 benchmarking runs that have been run so far on merge-commit 19fb7bb.

There was 1 benchmark result indicating a performance regression:

• Commit Run on amd64-m5-4xlarge-linux at 2025-06-24 11:06:04Z
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-07, scale_factor=1

The full Conbench report has more details.