fix(ci): remove cuda from all 8 ML sub-crate default features

The ml crate fix alone wasn't enough — ml-core, ml-dqn, ml-ppo,
ml-supervised, ml-ensemble, ml-explainability, ml-hyperopt, and
ml-labeling all had `default = ["cuda"]`, each independently pulling
in cudarc via candle-core/cuda.

Now `default = []` on all sub-crates. CUDA activates only when the
compile-and-train template passes `--features ml/cuda`, which
propagates through ml's cuda feature gate to all sub-crates.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

crates/ml-core/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "Shared ML types, traits, and infrastructure for Foxhunt"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda", "candle-nn/cuda"]
 s3-storage = ["aws-config", "aws-sdk-s3", "aws-types", "aws-credential-types", "urlencoding"]
 high-precision = ["rust_decimal/serde-float"]

crates/ml-dqn/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "DQN reinforcement learning for Foxhunt trading"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda", "candle-nn/cuda"]
 
 [dependencies]

crates/ml-ensemble/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "ML ensemble coordination — voting, confidence, gating, hot-swap, inference pipeline"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda"]
 
 [dependencies]

crates/ml-explainability/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "Model explainability (integrated gradients) for Foxhunt ML"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda", "candle-nn/cuda"]
 
 [dependencies]

crates/ml-hyperopt/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "ML hyperparameter optimization — PSO, TPE, campaigns, sensitivity analysis"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda"]
 
 [dependencies]

crates/ml-labeling/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "ML labeling algorithms for Foxhunt HFT training data"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda"]
 
 [dependencies]

crates/ml-ppo/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "PPO reinforcement learning for Foxhunt trading"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda", "candle-nn/cuda"]
 
 [dependencies]

crates/ml-supervised/Cargo.toml

@@ -14,7 +14,7 @@ categories.workspace = true
 description = "Supervised models (TFT, Mamba, Liquid, TGGN, TLOB, KAN, xLSTM, Diffusion)"
 
 [features]
-default = ["cuda"]
+default = []
 cuda = ["candle-core/cuda", "candle-nn/cuda"]
 
 [dependencies]