xlstream-eval: bound VLOOKUP and INDEX col index to range width

[cilladev]

Fixes: #143

Problem

builtin_vlookup and builtin_index extracted start_col from the table/range RangeRef but ignored end_col. A col_index/col_num past the range's last column read neighboring data from the loaded lookup sheet instead of returning #REF!. Example: =VLOOKUP(1, Lookup!A:A, 2, FALSE) returned column B's value instead of #REF!.

Fix

Both functions now capture end_col, compute the declared range width (end_col - start_col + 1), and return #REF! when the column index exceeds it — matching Excel. The check is skipped when end_col is None (whole-row refs already fail the start_col: Some(..) guard).

Changes

• crates/xlstream-eval/src/builtins/lookup.rs: capture end_col in VLOOKUP and INDEX destructures; add a width bound check returning #REF! for out-of-range column indices.
• Unit tests: vlookup_col_index_past_range_returns_ref, vlookup_col_index_at_range_width_ok, index_col_past_range_returns_ref.
• Conformance fixture issue-143-lookup-range-bounds.xlsx (Excel-populated): 38 formula cells across 5 rows covering valid lookups, #REF! (col past 1-col and 3-col ranges, both functions), plus #N/A (miss) and #VALUE! (col_index 0) singletons.

Scope note

The fix targets column bounds, as the issue describes. INDEX with a row past the loaded data of a whole-column range (INDEX(A:C, 100, 1)) is a separate concern: Excel treats it as in-range-empty (0) while xlstream returns #REF! (it only loads actual data rows). That deviation is pre-existing and intentional (see the existing index_out_of_bounds unit test), so it is out of scope here and excluded from the fixture.

Test plan

• Regression unit tests added and pass
• Conformance fixture created (Excel-populated) and passes
• cargo test -p xlstream-eval passes (2265 lib + 137 doc, no regressions)
• make check passes

Closes #143

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: bb31644	Previous: c90c6b8	Ratio
arithmetic/concat	258 ns/iter (± 13)	210 ns/iter (± 2)	1.23

This comment was automatically generated by workflow using github-action-benchmark.