pub fn wide_rows_enabled() -> boolExpand description
2026-09-25: The opt-in widening: true when METRALE_W4A16_TC_WIDE is set to a non-empty value
and tc_enabled. Read once per process. It raises narrow_gemv_max_rows to 16 and
lets tc_fixed_m route the fixed-M launchers (w4a16_gemv_batch2/3,
w4a16_gemv_dual_batch2/3) to the tensor-core kernels.