Merge pull request #320 from AdaWorldAPI/claude/fold-distillation-pr-wave-s57uj7

simd: keyed-reduction family — group count/min/max over resident or VIA keys

Author: AdaWorldAPI

src/simd.rs

@@ -826,6 +826,12 @@ pub use crate::simd_masking_ops::{
     mask_ternlog_assign,
     mask_xor,
     mask_xor_assign,
+    masked_group_count_u32,
+    masked_group_count_u32_via,
+    masked_group_max_i32,
+    masked_group_max_i32_via,
+    masked_group_min_i32,
+    masked_group_min_i32_via,
     masked_group_sum_i32,
     masked_group_sum_i32_via,
     masked_key_run_count_u32,