feat: add feature flags and scaffolding for WASM cognitive stack

Add canonical, spectral, cold-tier, and canonical-witness feature flags
across ruvector-mincut, ruvector-coherence, ruvector-gnn, and
cognitum-gate-kernel. Create ruvector-cognitive-container crate skeleton.

Implementation agents are building the full modules in parallel.

https://claude.ai/code/session_018QKTLyCUrMUQCRDqoiyEHY

crates/cognitum-gate-kernel/Cargo.toml

@@ -35,6 +35,7 @@ harness = false
 [features]
 default = ["std"]
 std = ["ruvector-mincut"]
+canonical-witness = []  # Canonical pseudo-deterministic witness fragments
 
 [profile.release]
 opt-level = "z"      # Optimize for size

crates/ruvector-cognitive-container/Cargo.toml

@@ -0,0 +1,25 @@
+[package]
+name = "ruvector-cognitive-container"
+version.workspace = true
+edition.workspace = true
+rust-version.workspace = true
+license.workspace = true
+authors.workspace = true
+repository.workspace = true
+description = "Verifiable WASM cognitive container with canonical witness chains"
+keywords = ["wasm", "cognitive", "container", "witness-chain", "deterministic"]
+categories = ["algorithms", "wasm", "cryptography"]
+
+[dependencies]
+serde = { workspace = true, features = ["derive"] }
+serde_json = { workspace = true }
+thiserror = { workspace = true }
+
+[dev-dependencies]
+proptest = { workspace = true }
+
+[features]
+default = []
+
+[lib]
+crate-type = ["rlib"]

crates/ruvector-cognitive-container/src/error.rs

@@ -0,0 +1,64 @@
+use thiserror::Error;
+
+#[derive(Error, Debug)]
+pub enum ContainerError {
+    #[error("Memory allocation failed: requested {requested} bytes, available {available}")]
+    AllocationFailed { requested: usize, available: usize },
+
+    #[error("Epoch budget exhausted: used {used} of {budget} ticks")]
+    EpochExhausted { used: u64, budget: u64 },
+
+    #[error("Witness chain broken at epoch {epoch}")]
+    BrokenChain { epoch: u64 },
+
+    #[error("Invalid configuration: {reason}")]
+    InvalidConfig { reason: String },
+
+    #[error("Container not initialized")]
+    NotInitialized,
+
+    #[error("Slab overflow: component {component} exceeded budget")]
+    SlabOverflow { component: String },
+}
+
+pub type Result<T> = std::result::Result<T, ContainerError>;
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_error_display() {
+        let err = ContainerError::AllocationFailed {
+            requested: 1024,
+            available: 512,
+        };
+        assert!(err.to_string().contains("1024"));
+        assert!(err.to_string().contains("512"));
+    }
+
+    #[test]
+    fn test_error_variants() {
+        let err = ContainerError::EpochExhausted {
+            used: 100,
+            budget: 50,
+        };
+        assert!(err.to_string().contains("100"));
+
+        let err = ContainerError::BrokenChain { epoch: 42 };
+        assert!(err.to_string().contains("42"));
+
+        let err = ContainerError::InvalidConfig {
+            reason: "bad value".to_string(),
+        };
+        assert!(err.to_string().contains("bad value"));
+
+        let err = ContainerError::NotInitialized;
+        assert!(err.to_string().contains("not initialized"));
+
+        let err = ContainerError::SlabOverflow {
+            component: "graph".to_string(),
+        };
+        assert!(err.to_string().contains("graph"));
+    }
+}

crates/ruvector-coherence/Cargo.toml

@@ -11,3 +11,7 @@ description = "Coherence measurement proxies for comparing attention mechanisms"
 [dependencies]
 serde = { workspace = true, features = ["derive"] }
 serde_json = { workspace = true }
+
+[features]
+default = []
+spectral = []  # Spectral coherence scoring for graph index health

crates/ruvector-coherence/src/lib.rs

@@ -9,9 +9,19 @@ pub mod comparison;
 pub mod metrics;
 pub mod quality;
 
+#[cfg(feature = "spectral")]
+pub mod spectral;
+
 pub use batch::{evaluate_batch, BatchResult};
 pub use comparison::{
     compare_attention_masks, edge_flip_count, jaccard_similarity, ComparisonResult,
 };
 pub use metrics::{contradiction_rate, delta_behavior, entailment_consistency, DeltaMetric};
 pub use quality::{cosine_similarity, l2_distance, quality_check, QualityResult};
+
+#[cfg(feature = "spectral")]
+pub use spectral::{
+    compute_degree_regularity, estimate_effective_resistance_sampled, estimate_fiedler,
+    estimate_largest_eigenvalue, estimate_spectral_gap, CsrMatrixView, HealthAlert,
+    HnswHealthMonitor, SpectralCoherenceScore, SpectralConfig, SpectralTracker,
+};

crates/ruvector-gnn/Cargo.toml

@@ -49,6 +49,7 @@ simd = []
 wasm = []
 napi = ["dep:napi", "dep:napi-derive"]
 mmap = ["dep:memmap2", "dep:page_size"]
+cold-tier = ["mmap"]  # Hyperbatch training for graphs exceeding RAM
 
 [dev-dependencies]
 criterion = { workspace = true }

crates/ruvector-mincut/Cargo.toml

@@ -42,7 +42,7 @@ mockall = { workspace = true }
 
 [features]
 default = ["exact", "approximate"]
-full = ["exact", "approximate", "integration", "monitoring", "simd", "agentic", "jtree", "tiered"]
+full = ["exact", "approximate", "integration", "monitoring", "simd", "agentic", "jtree", "tiered", "canonical"]
 exact = []  # Exact minimum cut algorithm
 approximate = []  # (1+ε)-approximate algorithm
 integration = ["ruvector-graph"]  # GraphDB integration
@@ -52,6 +52,7 @@ wasm = []  # WASM compatibility mode
 agentic = []  # 256-core parallel agentic chip backend
 jtree = []  # j-Tree hierarchical decomposition (ADR-002)
 tiered = ["jtree", "exact"]  # Two-tier coordinator (j-tree + exact)
+canonical = []  # Pseudo-deterministic canonical min-cut via cactus representation
 all-cut-queries = ["jtree"]  # Sparsest cut, multiway, multicut queries
 
 [lib]