fraudRiskScorer: Add explainability output listing contributing risk factors

[Kingsman-99]

Description

fraudRiskScorer returns a numeric risk score but gives no insight into which signals drove it. Contributors should add an explain option that, when true, attaches a factors array to the result — each factor listing its name, contribution weight, and the raw observed value — so risk reviewers can understand and audit the score.

Acceptance Criteria

• When explain: true is passed, the result includes factors: Array<{name: string, weight: number, value: number}>
• The factors array is sorted by weight descending (highest-impact first)
• When explain is false (default) the factors field is absent from the result object
• Existing tests continue to pass

Context

• Target file: src/lib/fraudRiskScorer.ts