feat(probes): add PII leakage probe

[cnaples79]

Summary

This PR adds a new probe to detect personal information (PII) leakage from LLMs. The probe is based on the paper "Extracting Training Data from Large Language Models" (https://arxiv.org/abs/2012.07805).

Changes

• Added a new probe garak.probes.personal.PII.
• Added a new detector garak.detectors.pii.ContainsPII.
• Added a new dataset garak/resources/pii.txt with examples of PII.
• Added tests for the new probe and detector.

Rationale

This probe helps to evaluate the risk of LLMs leaking sensitive personal information that may have been present in their training data.

Fixes #219

[github-actions[bot]]

DCO Assistant Lite bot All contributors have signed the DCO ✍️ ✅

[cnaples79]

I have read the DCO Document and I hereby sign the DCO

[cnaples79]

recheck

[leondz]

Thanks, will take a look!

[cnaples79]

@leondz Sounds goood! I'll address any issues if they come up.

[cnaples79]

@jmartin-tech thanks for the thorough review. I'm going to use your feedback and I'll update the PR.

Do you have any other feedback on how I could improve the PII examples? Or perhaps how to gather more relevant samples that would actually introduce risk?

[leondz]

bumped to draft until tests pass