feat(functions-nested): add array_filter higher-order function

[ologlogn]

Which issue does this PR close?

Partially addresses #14509 (Add Native Support for List Functions) — implements list_filter / array_filter.

Rationale for this change

array_transform was recently added as the first HigherOrderUDF (#21679). array_filter is the natural companion: filter elements of an array using a boolean lambda predicate, analogous to Spark's filter and DuckDB's list_filter.

What changes are included in this PR?

• New HigherOrderUDF ArrayFilter in datafusion/functions-nested/src/array_filter.rs
  • array_filter(array, x -> condition) with alias list_filter
  • Evaluates a boolean lambda per element; elements where the lambda returns true are kept
  • Null predicate results treated as false (element dropped), matching Spark semantics
  • Handles List and LargeList, sliced arrays, and null sublists correctly
  • Core logic: per-element boolean selection mask + recomputed per-sublist offsets via arrow::compute::filter
• Shared HOF helpers extracted to macros_lambda.rs (value_lambda_pair, coerce_single_list_arg) to avoid duplication with array_transform
• Shared test utilities in test_utils.rs (create_i32_list, eval_hof_on_i32_list, v())
• Registered in all_default_higher_order_functions() and expr_fn

Are these changes tested?

Yes, unit tests cover:

• Basic filtering (x -> x > 2)
• Multiple sublists with different per-sublist filter counts
• Sliced list arrays (unreachable values not evaluated)
• Null sublists (predicate not evaluated on null sublist values; null preserved in output)
• All-filtered-out (empty output sublist)

Are there any user-facing changes?

Yes — array_filter(array, lambda) and its alias list_filter(array, lambda) are now available as SQL functions.

[ologlogn]

Hi @gabotechs, could you please trigger CI? Thanks!