Remove the manual application option for hints yaml

[melonattacker]

This pull request removes support for user-supplied YAML hints ("hint.yaml") from Threat Thinker, shifting the workflow to rely solely on LLM-inferred attribute inference and business context for diagram analysis. Documentation, CLI, code, and examples have been updated to reflect this change, and references to hints and hint files have been removed throughout the project.

Major changes include:

Removal of Hint YAML Support

• Deleted the docs/hints.md guide, all references to "hint.yaml", and the --hints CLI argument and processing logic from the codebase, including the apply_hints function and related UI steps. Example hint files in examples/diagrams/web/hint.yaml and examples/diagrams/aws/hint.yaml were also removed. [1] [2] [3] [4] [5] [6] [7] [8]

Documentation and Tutorial Updates

• Updated documentation (docs/README.md, docs/diagram.md, docs/design.md, docs/cli.md, docs/troubleshooting.md, docs/tutorials.md) to remove or revise all mentions of hints, hint files, and hint-related CLI flags. Guidance now emphasizes using LLM-inferred attributes and business context for customizing analysis. [1] [2] [3] [4] [5] [6] [7] [8]

CLI and Example Cleanup

• Removed the --hints argument and related examples from CLI help, usage, and code (src/threat_thinker/main.py). Progress tracking and UI steps were adjusted to reflect the new workflow without hint application. (F8cbae7fL13R13, [1] [2] [3]

Dependency Update

• Changed the YAML parsing dependency from pyaml to PyYAML in pyproject.toml for consistency and standardization.

Codebase Refactoring

• Removed the apply_hints function and all imports/usages of it from src/threat_thinker/hint_processor.py and src/threat_thinker/main.py. The code now exclusively uses LLM-inferred attribute merging. [1] [2] F8cbae7fL55R55)

These changes streamline the user experience and codebase by focusing on LLM-powered attribute inference and business context, eliminating the need for manual YAML hint files.