chore: prepare 0.0.14 release

[MattyB95]

Summary

• Adds in-process caching for every model family (HuBERT, Wav2Vec2, WavLM, AST, VIT, Spectra0, SpectraAASIST, SpectraAASIST3, RawNet2, Classical) via a new Utilities/model_cache.py helper, so repeated detect() calls for the same model reuse already-loaded weights instead of reloading them from disk every time
• Adds direct regression tests for the new cache (tests/test_model_cache.py) plus a "loaded once across repeated calls" test per caching integration point
• Expands the ruff rule set with UP, B, SIM, and RUF and fixes the resulting findings across the first-party codebase; Models/RawNet2/model.py is exempted since it's vendored third-party code
• Updates CONTRIBUTING.md's "adding a new model" template, which predated the run_pipeline() consolidation and would have produced duplicate, uncached code
• Updates all dependencies to their latest compatible versions and raises pyproject.toml's floors to match (notably huggingface-hub 1.14->1.22, transformers 5.8->5.13, scikit-learn 1.8->1.9, ty 0.0.35->0.0.56)
• Fixes a UnicodeEncodeError crash in the CLI on non-UTF-8 consoles (e.g. Windows cp1252, the default outside Windows Terminal/UTF-8 mode) - DetectionResult.__str__ embeds checkmark/cross emoji, so main() now reconfigures stdout/stderr to UTF-8
• Bumps the version to 0.0.14 and updates the changelog

Test plan

• just check passes locally (lint, format, type-check)
• just test passes locally (158 tests, 99% coverage)
• uv audit reports no known vulnerabilities
• Docs build (mkdocs build --strict) passes locally
• CI passes on this PR (lint, type-check, all Python versions)
• After merge to main, confirm the release workflow tags v0.0.14
• Confirm the GitHub Release body matches the 0.0.14 changelog section

[Copilot]

Pull request overview

Prepares the 0.0.14 release by introducing an in-process model/pipeline cache (to avoid repeated on-disk reloads across repeated detect() calls), adding regression tests around caching behavior, and updating release/versioning + dependency/tooling configuration.

Changes:

• Introduce a shared cached_loader() / clear_all() cache registry and integrate it into model/pipeline loading paths.
• Add regression tests validating “loaded once across repeated calls” behavior and ensure caches are cleared between tests.
• Bump version/docs/changelog and refresh dependency floors + CI workflow pins.

Reviewed changes

Copilot reviewed 35 out of 36 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/test_models.py	Adds “loaded once across repeated calls” regression tests and refactors patch nesting.
tests/test_model_cache.py	New unit tests for cache hit/eviction and clear_all() behavior.
tests/test_jabberjay.py	Refactors nested context managers and adds non-UTF8 stdout CLI regression test.
tests/test_architecture.py	Adjusts unused variables to satisfy expanded linting.
tests/conftest.py	Adds an autouse fixture to clear all model caches between tests.
src/Jabberjay/Utilities/pipeline.py	Caches transformers audio pipelines via a new _load_pipeline() loader.
src/Jabberjay/Utilities/model_cache.py	Introduces shared cached_loader() decorator + global registry + clear_all().
src/Jabberjay/Utilities/enum_handler.py	Modernizes super() usage.
src/Jabberjay/Models/Transformer/VIT/MFCC/run.py	Caches VIT MFCC image-classification pipeline.
src/Jabberjay/Models/Transformer/VIT/MelSpectrogram/run.py	Caches VIT MelSpectrogram image-classification pipeline.
src/Jabberjay/Models/Transformer/VIT/ConstantQ/run.py	Caches VIT ConstantQ image-classification pipeline.
src/Jabberjay/Models/SpectraAASIST3/run.py	Switches to cached shared Spectra loader.
src/Jabberjay/Models/SpectraAASIST/run.py	Switches to cached shared Spectra loader.
src/Jabberjay/Models/Spectra0/run.py	Switches to cached shared Spectra loader.
src/Jabberjay/Models/Spectra0/model.py	Minor simplification from expanded lint rules.
src/Jabberjay/Models/Spectra/shared.py	Adds cached load_pretrained() helper used by Spectra model families.
src/Jabberjay/Models/RawNet2/run.py	Refactors RawNet2 loading into a cached _load_model(device) path.
src/Jabberjay/Models/Classical/run.py	Refactors joblib classifier loading into cached _load_classifier().
src/Jabberjay/jabberjay.py	Adds UTF-8 stdout/stderr reconfiguration and improves exception chaining.
src/Jabberjay/init.py	Reorders __all__.
README.md	Documents in-process caching and bumps referenced version to 0.0.14.
pyproject.toml	Bumps version/dependency floors; expands ruff rules and adds per-file ignores.
examples/run_all.py	Updates comment text to avoid ambiguous unicode.
examples/quickstart.py	Updates range comment formatting to avoid ambiguous unicode.
examples/exploring_results.py	Updates range comment formatting to avoid ambiguous unicode.
docs/models.md	Documents new in-process caching behavior.
docs/getting-started.md	Documents new in-process caching behavior.
CONTRIBUTING.md	Updates “add a new model” template to use cached loading patterns.
CITATION.cff	Bumps version/date-released for 0.0.14.
CHANGELOG.md	Adds 0.0.14 release notes + comparison link.
.github/workflows/sync-develop.yml	Bumps actions/checkout major version.
.github/workflows/release.yml	Bumps actions/checkout major version.
.github/workflows/docs.yml	Bumps actions/checkout major version.
.github/workflows/codeql.yml	Bumps actions/checkout major version.
.github/workflows/ci.yml	Bumps actions/checkout major version; updates publish permissions/config.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.