feat: faster loading if model already cached

[stephantul]

This PR adds much faster model loading by directly loading from cache if a model is cached, if force_download is set to true during loading

Timings when loading potion-base-8m

• Before: 1.27 s ± 72.5 ms per loop
• Now: 37.3 ms ± 335 µs per loop

This speed is only enabled when setting force_download to False. It is true by default for backwards compatibility and download tracking

Mention for #275

[codecov]

Codecov Report

❌ Patch coverage is 66.66667% with 9 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
model2vec/hf_utils.py	62.50%	9 Missing ⚠️

Files with missing lines	Coverage Δ
model2vec/distill/inference.py	97.46% <100.00%> (ø)
model2vec/model.py	93.70% <100.00%> (-0.30%)	⬇️
model2vec/utils.py	90.74% <ø> (-1.80%)	⬇️
model2vec/hf_utils.py	76.23% <62.50%> (+1.23%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.