feat: add metadata-driven k-mer filtering for rebuild command

Introduces a metadata-driven filtering system for the rebuild command, classifying genomes into ingroup and outgroup categories using exact, inequality, and hierarchical path predicates. Implements a GroupQuorumFilter to enforce configurable presence thresholds and fraction constraints per group. Refactors the command to replace global quorum filters with this unified approach, converts the presence flag to a threshold parameter, and adds corresponding documentation and MkDocs navigation.

src/obikpartitionner/src/filter.rs

@@ -85,3 +85,52 @@ impl KmerFilter for MaxTotalCount {
         row.iter().sum::<u32>() <= self.total
     }
 }
+
+// ── Group-based quorum filter ─────────────────────────────────────────────────
+
+/// Quorum filter operating on pre-classified genome groups.
+///
+/// `ingroup_idx` / `outgroup_idx` are column indices into the per-genome row.
+/// When `ingroup_idx` is empty, no ingroup quorum is checked.
+/// When `outgroup_idx` is empty, no outgroup quorum is checked.
+pub struct GroupQuorumFilter {
+    pub ingroup_idx:        Vec<usize>,
+    pub outgroup_idx:       Vec<usize>,
+    pub threshold:          u32,
+    pub min_count:          usize,
+    pub max_count:          usize,
+    pub min_frac:           f64,
+    pub max_frac:           f64,
+    pub min_outgroup_count: usize,
+    pub max_outgroup_count: usize,
+    pub min_outgroup_frac:  f64,
+    pub max_outgroup_frac:  f64,
+}
+
+impl KmerFilter for GroupQuorumFilter {
+    fn passes(&self, row: &[u32], _n_genomes: usize) -> bool {
+        if !self.ingroup_idx.is_empty() {
+            let n = self.ingroup_idx.iter()
+                .filter(|&&i| row.get(i).copied().unwrap_or(0) > self.threshold)
+                .count();
+            let denom = self.ingroup_idx.len();
+            if n < self.min_count { return false; }
+            if n > self.max_count { return false; }
+            let frac = n as f64 / denom as f64;
+            if frac < self.min_frac { return false; }
+            if frac > self.max_frac { return false; }
+        }
+        if !self.outgroup_idx.is_empty() {
+            let n = self.outgroup_idx.iter()
+                .filter(|&&i| row.get(i).copied().unwrap_or(0) > self.threshold)
+                .count();
+            let denom = self.outgroup_idx.len();
+            if n < self.min_outgroup_count { return false; }
+            if n > self.max_outgroup_count { return false; }
+            let frac = n as f64 / denom as f64;
+            if frac < self.min_outgroup_frac { return false; }
+            if frac > self.max_outgroup_frac { return false; }
+        }
+        true
+    }
+}

src/obikpartitionner/src/lib.rs

@@ -8,6 +8,6 @@ mod partition;
 mod query_layer;
 mod rebuild_layer;
 
-pub use filter::KmerFilter;
+pub use filter::{GroupQuorumFilter, KmerFilter};
 pub use merge_layer::MergeMode;
 pub use partition::{KmerPartition, KmerSpectrum, PARTITIONS_SUBDIR};