refactor(ml): delete mixed_precision module — BF16 unconditional on CUDA

Eliminate the entire mixed_precision runtime indirection layer:
- Delete crates/ml-core/src/mixed_precision.rs (training_dtype, ensure_training_dtype, align_dim_for_tensor_cores)
- Inline ~100 call sites across 130 files to constants:
  training_dtype(&device) → candle_core::DType::BF16
  ensure_training_dtype(x) → x.to_dtype(candle_core::DType::BF16)
  align_dim_for_tensor_cores(x, &device) → (x + 7) & !7
- Remove re-exports from ml-dqn, ml-supervised, ml lib.rs
- Clean config/toml/json/shell references

No CPU/Metal training path exists — BF16 is the only dtype.

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

scripts/gpu-hotpath-guard.sh

@@ -33,7 +33,6 @@ HOT_PATHS=(
     "gpu_backtest_evaluator"
     "gpu_training_guard"
     "gpu_weights"
-    "mixed_precision.rs"
     # DQN crate — all modules (inference, networks, replay, regularization)
     "ml-dqn/src/"
     # PPO training + inference