Add scoring doc, update readme

[omkar-foss]

Description
Add scoring doc, update readme

This PR closes #66, #69. By addition of SCORING.md, it potentially also resolves #49.

Notes for Reviewers

Signed commits

• Yes, I signed my commits.

Generative AI disclosure

Please select one option:

• This contribution was NOT assisted or created by Generative AI tools.
• This contribution was assisted or created by Generative AI tools.

If AI tools were used, please provide details below:
- What tools were used?
- How were these tools used?
- Did you review these outputs before submitting this PR?

[andrew]

• The confidence ranges in SCORING.md do not match the implementation. A score of 0 has confidence none, not low, and scores above 100 remain high. Please document the ranges as none = 0, low = 1–30, medium = 31–70, and high = 71+. The related issue link on line 3 also points to the old repository and returns 404.

• Issue #66 asks for CLI examples and this PR says it closes that issue, but SCORING.md only contains a commit example. Please add examples showing the default scoring output and how to use --confidence-levels.

• The detector summary says both Co-Authored-By and Assisted-By require known AI tool emails. Only Co-Authored-By checks known emails. Assisted-By accepts any non-empty tool name, so please describe these separately.

[MoralCode]

Why do we have "low/med/high" if we are also providing the numerical value?

[omkar-foss]

We've had a filter report by confidence feature (this one) which depends on confidence

[MoralCode]

additive scoring is subject to drift as new detection methods are added, especially if this is the numeric system we are exposing to users and downstream applications that expect stability

[omkar-foss]

I had added weight average scoring as discussed here, but we settled on additive scoring (SpamAssassin style), see this review.