Clean structure

[srmnitc]

Rules

• All classes are fully fledged pydantic objects
• All classes should serialise and return themselves to their top level object
• Serialisation can be called automatically or manually - decision needed
• All data properties are fields
• All classes should unserialise themselves
• Should classes include reference to their top level object - decision needed
• Named Individuals - do they create instance of themselves and assign as type - or no?

TODO

• Make sure that to_graph method called on Activity does not create structures again
• Implement Activity method on transformation
• Create a new Workflow object and include parsing
• Define wrappers for wf envs
• update tests
• update examples
• what about substitute and interstitial?
• check translate and shear and differences if needed
• improvement to the 'id' field - if provided as user input - search kg and update existing record
• Add PIDs everywhere
• Unit cell angles should be recorderd
• Based on unit cell angles, the cell should be either primitive or cubic or check for more options

---

This pull request introduces several enhancements and refactors across the atomrdf library, focusing on modularizing the codebase, improving functionality for crystal structure generation, and standardizing I/O operations. Key changes include the addition of new utility functions, restructuring of I/O operations, and the introduction of new plotting and transformation capabilities.

New Features and Enhancements:

Crystal Structure Generation:

• Added bulk and lattice functions in atomrdf/build/bulk.py to create and manage crystal structures with customizable parameters such as lattice constants, element types, and noise. These functions integrate with pyscal3 and atomrdf systems for seamless structure creation.

I/O Operations:

• Introduced a new write function in atomrdf/io/write.py to handle file exports in various formats, including LAMMPS, POSCAR, and Quantum Espresso. This replaces the older _write_espresso implementation, offering better modularity and support for additional formats.
• Added a read function in atomrdf/io/read.py to facilitate importing structures from multiple file formats, with options for specifying lattice properties and repetitions.

Plotting Capabilities:

• Added plot and plot3d functions in atomrdf/io/plot.py for visualizing atomic structures. These functions support multiple styles, such as property-based coloring and boolean property visualization, leveraging pyscal3 and pyiron_atomistics.

Code Refactoring:

Modularization:

• Refactored imports and utilities across the codebase for better organization and maintainability. For example:
  • Moved plotting and I/O utilities into dedicated modules (atomrdf/io/plot.py, atomrdf/io/write.py, atomrdf/io/read.py). [1] [2]
  • Simplified atomrdf/graph.py by replacing direct calls to to_file with the new write function.

Removal of Deprecated Code:

• Removed the old _convert_tab_to_dict and write_espresso implementations from atomrdf/io.py, consolidating their functionality into the new write module.

Minor Updates:

• Added utility function _declass in atomrdf/build/buildutils.py to standardize handling of special variables like Property objects.
• Updated atomrdf/transform/__init__.py to re-export all transformations for easier access.