fix: model import

Merge main into huggingface, retain HF frontmatter

chore: full sync of claude-forge skill files

chore: add pre-planning context gathering to planner agent

chore: sync skills to latest claude-forge

ci: fix auto-merge workflow flags and merge strategy

ci: enforce CI gate and skip major updates in dependabot auto-merge

ci: add dependabot auto-merge workflow

build(deps): bump actions/checkout from 4 to 6 (#6)

build(deps): bump actions/setup-python from 5 to 6 (#7)

ci: add concurrency groups, timeouts, and path filtering

chore: sync skills from claude-forge with manifest logging

chore: add skill-runs manifest for tracking claude-forge skill usage

Merge pull request #4 from HatmanStack/audit-update

fix: address critical and medium post-audit findings

fix: address post-audit review findings across docs and source

docs: add audit plan artifacts and project roadmap

chore(docs): mark Phase-5 tasks complete

docs(readme): document data file paths and configuration

docs(readme): document training script prerequisites

docs(readme): update installation to use pyproject.toml and uv

docs(readme): update project structure tree to match codebase

docs(readme): fix feature descriptions and data source terminology

chore(config): remove obsolete ruff ignore rules

ci(coverage): enforce 70% coverage threshold in CI

refactor(types): verify type annotations are already tightened

ci(hooks): add pre-commit config with ruff and mypy hooks

ci(deps): consolidate to pyproject.toml, remove requirements files

docs: mark Phase 3 tasks complete

style(state): remove unused pytest import from test_state

test(coverage): raise coverage threshold from 50% to 70%

test(ml): add integration test with real model file

test(utils): add tests for HTML escaping utilities

test(state): add tests for session state management functions

test(database): add integration test for CSV column validation

docs: mark Phase 2 tasks complete

fix(config): move logging setup from module-level to explicit init

fix(scripts): replace destructive del with slicing in create_stats

fix(models): remove duplicate validation in DifficultySettings

fix(ml): add input shape validation before model prediction

fix(logging): replace f-string logging with lazy %s formatting

refactor(ml): decouple Streamlit caching from model module

refactor(database): decouple Streamlit caching from connection module

refactor(pages): extract shared configure_page, remove finally:pass

refactor(validation): remove SQL injection guards (no SQL database exists)

chore(config): remove debug_streamlit.py from ruff per-file-ignores

refactor(dead-code): remove unused PlayerStats Pydantic model

refactor(dead-code): remove unused get_player_by_full_name and safe_styled_text

refactor(dead-code): remove unused GameState and 4 dead functions from session.py

chore(gitignore): expand .gitignore to cover build artifacts and caches