GH-49958: Do not merge [C++][Dataset] DEBUG -> blocked on upstream gcc-16 fix

[tadeja]

Rationale for this change

Debug #49958

What changes are included in this PR?

Only debug for now.

Are these changes tested?

CI debug

Are there any user-facing changes?

No

• GitHub Issue: [C++][CI] MinGW Gcc 15.2->16.1 toolchain regression - memory corruption in multi-threaded tests #49958

[github-actions]

⚠️ GitHub issue #49958 has been automatically assigned in GitHub to PR creator.

[tadeja]

Both failing tests throw bad_weak_ptr at the same line dataset_writer.cc:212 DatasetWriterFileQueue::ScheduleBatch:

  void ScheduleBatch(std::shared_ptr<RecordBatch> batch) {
    file_tasks_->AddSimpleTask(
        [self = SfwDbg(__LINE__) /* was shared_from_this() */, batch = std::move(batch)]() {
          return self->WriteNext(std::move(batch));
        },
        "DatasetWriter::WriteBatch"sv);
  }

https://github.com/apache/arrow/actions/runs/25690826182/job/75426470835#step:12:614

[ RUN      ] DatasetWriterTestFixture.BatchWriteConcurrent
GH-49958: bad_weak_ptr DatasetWriterFileQueue dataset_writer.cc:212
terminate called after throwing an instance of 'std::bad_weak_ptr'
  what():  bad_weak_ptr

      Start 62: arrow-dataset-dataset-writer-test
    Test #62: arrow-dataset-dataset-writer-test ............Exit code 0xc0000374***Exception:   0.34 sec

https://github.com/apache/arrow/actions/runs/25690826182/job/75426470835#step:12:794

[ RUN      ] TestFileSystemDataset.MultiThreadedWritePersistsOrder
GH-49958: bad_weak_ptr DatasetWriterFileQueue dataset_writer.cc:212
terminate called after throwing an instance of 'std::bad_weak_ptr'
  what():  bad_weak_ptr

      Start 65: arrow-dataset-file-test
    Test #65: arrow-dataset-file-test ......................***Exception: SegFault  0.20 sec

[tadeja]

Diagnosed - The make_shared did not fix this, and the new CI log shows:
https://github.com/apache/arrow/actions/runs/25731999852/job/75559307182?pr=49961#step:12:870

[ RUN      ] DatasetWriterTestFixture.BatchWriteConcurrent
GH-49958: ScheduleBatch this=000002f3e2fda080 EXPIRED use_count=0
GH-49958: bad_weak_ptr DatasetWriterFileQueue dataset_writer.cc:218
terminate called after throwing an instance of 'std::bad_weak_ptr'
  what():  bad_weak_ptr

(dataset_writer.cc:218 is the same as the previous :212 line - because of added diagnostics shifted by 6 lines)

(1) It's not the reset vs make_shared wiring
(2) By the time ScheduleBatch runs, the control block of file queue already reports use_count=0.

[tadeja]

Diagnosed further - this isn't a dataset_writer bug. It looks like lower-level memory corruption, likely a gcc-16/MinGW regression. Keeping workaround of #49931's MSYS2 pin for now, #49945 stays waiting, tracked under #49948.

• The bad_weak_ptr isn't an Arrow lifetime bug: in BatchWriteConcurrent the queue's only shared_ptr is not reset until Finish(), so weak_from_this().expired() there means a corrupted control block. reset -> make_shared did not fix this.
• adding fprintf to the constructor/destructor of DatasetWriterFileQueue made it a Heisenbug - failed once, retry passed OK:
  https://github.com/apache/arrow/actions/runs/25737251706/job/75577465287?pr=49961#step:12:604

    Test #62: arrow-dataset-dataset-writer-test ............Exit code 0xc0000374***Exception:   0.30 sec
...
    Test #62: arrow-dataset-dataset-writer-test ............   Passed    4.85 sec
...
The following tests FAILED:
	 41 - arrow-async-utility-test (Exit code 0xc0000374)   arrow-tests unittest
	 44 - arrow-threading-utility-test (Timeout)            arrow-tests unittest
	 65 - arrow-dataset-file-test (Failed)                  arrow_dataset unittest
	 78 - arrow-flight-test (Failed)                        arrow_flight unittest
Error: Process completed with exit code 8.

• Independent reproducers without dataset_writer are: arrow-async-utility-test always dies at PushGenerator.Stress and arrow-threading-utility-test always times out. Both fine on CLANG64/libc++ and on gcc‑15.2. Only gcc 15.2->gcc 16.1 is affected.