feat: add parallel k-mer counting and stats CLI

Introduces allocation-free `sum()` and `count_nonzero()` methods for compact integer vectors, extending the `ColumnWeights` trait with `partial_kmer_counts`. Adds parallel partition scanning to the k-mer index for computing per-genome distinct k-mer counts, and exposes a new `--stats` CLI flag to output these statistics as CSV.

src/obicompactvec/src/intmatrix.rs

@@ -54,6 +54,14 @@ impl ColumnarCompactIntMatrix {
         Array1::from_vec(sums)
     }
 
+    pub(crate) fn count_nonzero(&self) -> Array1<u64> {
+        let counts: Vec<u64> = (0..self.n_cols())
+            .into_par_iter()
+            .map(|c| self.col(c).count_nonzero())
+            .collect();
+        Array1::from_vec(counts)
+    }
+
     pub(crate) fn partial_bray_dist_matrix(&self) -> Array2<u64> {
         self.pairwise_u64(|i, j| self.col(i).partial_bray_dist(self.col(j)))
     }
@@ -234,6 +242,14 @@ impl PackedCompactIntMatrix {
         )
     }
 
+    pub(crate) fn count_nonzero(&self) -> Array1<u64> {
+        Array1::from_vec(
+            (0..self.n_cols).into_par_iter()
+                .map(|c| (0..self.n_rows).filter(|&s| self.get(c, s) > 0).count() as u64)
+                .collect()
+        )
+    }
+
     // ── Pair primitives ───────────────────────────────────────────────────────
 
     fn pair_partial_bray(&self, i: usize, j: usize) -> u64 {
@@ -421,6 +437,10 @@ impl PersistentCompactIntMatrix {
         match self { Self::Columnar(m) => m.sum(), Self::Packed(m) => m.sum() }
     }
 
+    pub fn count_nonzero(&self) -> Array1<u64> {
+        match self { Self::Columnar(m) => m.count_nonzero(), Self::Packed(m) => m.count_nonzero() }
+    }
+
     pub fn partial_bray_dist_matrix(&self) -> Array2<u64> {
         match self { Self::Columnar(m) => m.partial_bray_dist_matrix(), Self::Packed(m) => m.partial_bray_dist_matrix() }
     }
@@ -451,6 +471,7 @@ use crate::traits::{ColumnWeights, CountPartials};
 
 impl ColumnWeights for PersistentCompactIntMatrix {
     fn col_weights(&self) -> Array1<u64> { self.sum() }
+    fn partial_kmer_counts(&self) -> Array1<u64> { self.count_nonzero() }
 }
 
 impl CountPartials for PersistentCompactIntMatrix {

src/obicompactvec/src/reader.rs

@@ -133,11 +133,15 @@ impl PersistentCompactIntVec {
     }
 
     #[inline]
-    /// Returns the sum of all values in the compact int vector.
     pub fn sum(&self) -> u64 {
         self.iter().map(|v| v as u64).sum()
     }
 
+    #[inline]
+    pub fn count_nonzero(&self) -> u64 {
+        self.iter().filter(|&v| v > 0).count() as u64
+    }
+
     #[inline]
     /// Returns the Bray-Curtis distance between two compact int vectors.
     pub fn bray_dist(&self, other: &PersistentCompactIntVec) -> f64 {

src/obicompactvec/src/traits.rs

@@ -2,8 +2,16 @@ use ndarray::{Array1, Array2};
 
 /// Column-level weight statistic — total count or presence count per column.
 /// Additive across layers and partitions; used as denominator in normalised distances.
+///
+/// `partial_kmer_counts` returns the number of **distinct k-mers** present per
+/// column (presence = 1 entries; count > 0 entries). For presence matrices this
+/// equals `col_weights`; for count matrices it differs (count_nonzero vs sum).
 pub trait ColumnWeights: Send + Sync {
     fn col_weights(&self) -> Array1<u64>;
+
+    fn partial_kmer_counts(&self) -> Array1<u64> {
+        self.col_weights()
+    }
 }
 
 /// Partial distance matrices for count-based data (`PersistentCompactIntMatrix`).