adapt readme

[AnFreTh]

This pull request includes updates to the README.md file and changes to the pyproject.toml file to reflect the new version and dependencies. The most important changes include the addition of new features, updates to preprocessing methods, and modifications to the custom model implementation.

Documentation Updates:

• Added a "What's New" section to highlight new features such as individual preprocessing, latent representation extraction, and custom training metrics.
• Updated the preprocessing section to specify default methods and individual preprocessing methods for each feature.
• Added information on using pre-trained models for categorical encoding.
• Expanded the usage examples to include getting latent representations and using unstructured data.

Custom Model Implementation:

• Modified the custom model example to use a single feature_information parameter instead of separate categorical and numerical feature info.
• Updated the custom model class to include an embedding layer and handle feature information more efficiently.

Version and Dependency Updates:

• Updated the version from 1.1.0 to 1.2.0 in pyproject.toml.
• Adjusted the Python version compatibility and updated the torch dependency range.

[AnFreTh]

@mkumar73 I am not quite sure about the new workflow here. Poetry is breaking after the Torch version change. Not sure what's wrong. Can you take a look?

[mkumar73]

@AnFreTh - The issue occurred because the poetry.lock file was not updated after the Torch version was upgraded in the pyproject.toml. Running the poetry lock command fixed the problem, and the pull request was merged into the develop branch.