feat(training): add burn MLP and CART tree trainers with weight export

Behind the `training` feature flag (burn 0.20 + ndarray + autodiff).
Trains a single-hidden-layer MLP with Adam optimizer and weighted BCE
loss, plus a CART decision tree using Gini impurity. Exports trained
weights as Rust const arrays that compile directly into the binary.

Signed-off-by: Sienna Meridian Satterwhite <sienna@sunbeam.pt>

Cargo.toml

@@ -77,6 +77,17 @@ iroh-gossip = { version = "0.96", features = ["net"] }
 blake3      = "1"
 hex         = "0.4"
 rand        = "0.9"
+rayon       = "1"
+tempfile    = "3"
+
+# Dataset ingestion (CIC-IDS2017 CSV parsing)
+csv         = "1"
+
+# burn-rs ML framework (training only, behind `training` feature)
+burn = { version = "0.20", features = ["ndarray", "autodiff"], optional = true }
+
+[features]
+training = ["burn"]
 
 [dev-dependencies]
 criterion = { version = "0.5", features = ["html_reports"] }
@@ -87,6 +98,10 @@ tempfile  = "3"
 name = "scanner_bench"
 harness = false
 
+[[bench]]
+name = "ddos_bench"
+harness = false
+
 [profile.release]
 opt-level = 3
 lto       = true