Cleaning code

[jvsguerra]

This pull request introduces several updates to the codebase, focusing on configuration simplification, improved logging, and code cleanup. The most significant changes are the restriction of certain configuration options to simplify usage, refactoring of logging for consistency, and removal or modification of unused or redundant code to improve maintainability.

Configuration simplification

• Restricted the dssp_acc_array and rsa_method options in GraphConfig and related functions to only allow "Wilke" and "dssp" respectively, removing support for other values. This reduces complexity and potential for misconfiguration. [1] [2] [3]
• Removed usage of the rsa_table setting from the graph creation process in create_graphs, aligning with the new configuration restrictions.

Logging and output improvements

• Refactored logging setup and usage throughout the codebase for consistency, including import order fixes, improved verbosity handling, and removal of redundant debug statements. Also improved logging messages in chunk processing and step execution to be more informative and user-friendly. [1] [2] [3] [4] [5]
• Updated summary and debug logging in triad finding and processing functions to include protein names and chunk information, making logs easier to interpret during execution. [1] [2]

Code cleanup and refactoring

• Removed the unused remove_water_from_pdb function and related water-removal logic from preprocessing.py, simplifying the graph creation workflow and ensuring that original files are used directly. [1] [2]
• Refactored imports in several files for clarity and correctness, especially in pdb_graph_builder.py and preprocessing.py. [1] [2]
• Updated function signatures and variable usage to improve clarity and type safety, such as in get_exposed_residues and create_graphs. [1] [2]

These changes collectively streamline configuration, improve logging for debugging and monitoring, and remove unnecessary code for better maintainability.