v0.6.0 - Add LSTM Layer Implementation

Release Note for RustyML v0.6.0

Added

• Add LSTM layer implementation
• Add get_weights method across layers and LayerWeight enum
• Ensure fit validates optimizer and layers before training
• Add detailed documentation and examples to neural network layers
• Add regularization support to linear and logistic regression

Changed

• Refactor KernelType into machine_learning module
• Refactor regressors to use shared RegressorCommonGetterFunctions trait
• Refactor LinearSVC to use RegularizationType for penalties
• Refactor traits in neural_network module definitions into traits module

Removed

• Remove method update_parameters in Layer trait
• Remove getter methods from Dense layer