refactor: improve code formatting and organization across multiple API and benchmark files

fix: correct dataset loading in DrawBenchPrompts to ensure proper retrieval of test data

fix: update dataset loading in DrawBenchPrompts to retrieve full test set

chore: update .gitignore, environment.yml, and README; remove nils_installs.txt

feat: add aws canva + examples in readme + my current environment + disable HPS for now

fix: fix several erros

feat: add core logic for project