docs : clarify unused test utility

[Nachiket-Roy]

Which issue does this PR close?

Closes N/A (no existing issue)
This change addresses a compiler warning observed during local test runs and does not correspond to a tracked bug or feature request.

Rationale for this change

During local test runs, the helper function ctx_and_codec in datafusion/ffi/tests/utils triggered a dead_code warning.
While the function is currently unused, it encapsulates non-trivial setup logic for FFI integration tests and is expected to be reused as FFI test coverage expands.
This change documents the intent of the helper and explicitly marks it as intentionally retained scaffolding, improving code clarity.

What changes are included in this PR?

• Added documentation explaining the purpose of the ctx_and_codec helper
• Explicitly marked the helper as intentionally unused to silence dead_code warnings

Are these changes tested?

Yes.

• This change is documentation-only and does not affect runtime behavior.
• Existing test coverage remains unchanged and continues to pass.

Are there any user-facing changes?

No.
This PR only affects internal test utilities and does not impact public APIs or user-facing behavior.

[Jefffrey]

Suggested change

	#[allow(dead_code)]
	#[expect(dead_code)]

• Thanks for this, after chore: enforce clippy::allow_attributes for datafusion-ffi crate #19480 we'll need to change this to an expect

[Nachiket-Roy]

Thanks for the review. Updated to #[expect(dead_code)].

[Nachiket-Roy]

The lint expectation is currently unfulfilled so #[expect(dead_code)] fails CI.
Should I revert this back to #[allow(dead_code)] for now and switch to #[expect] once the lint actually triggers?

[Jefffrey]

We should figure out why it isn't fulfilled in that case; it doesn't make sense that allow suppresses a lint that isn't fulfilled if we replace it with expect unless its a bug in the linter. Perhaps check if it occurs when a certain feature is enabled/disabled?

[Nachiket-Roy]

The helper is used by FFI integration tests, but its usage depends on which tests and features are compiled. In some configurations, the compiler does not see any references to it and therefore does not emit a dead_code lint. In those cases, #[expect(dead_code)] fails because the lint is not triggered, while #[allow(dead_code)] still works since it does not require the lint to be emitted.

[Nachiket-Roy]

I’ve switched this back to #[allow(dead_code)] for now. Since the dead_code lint isn’t consistently emitted across all test / feature configurations, using #[expect(dead_code)] causes failures when the lint doesn’t trigger. allow avoids that while still documenting the intent.

[Jefffrey]

#[cfg_attr(not(feature = "integration-tests"), expect(dead_code))]

This should work; it conditionally expects this as dead code if not using integration-tests feature

[Nachiket-Roy]

Alright. Thanks.