Sources that always work (botvrij, disconnect, oisd, openphish, phishtank, quad9) drop their user-facing enable_* option; the rule's enabled/disabled state is now solely controlled by the SDK rule toggle. Sources that require credentials (criminalip, malwarebazaar, otx, pulsedive, safebrowsing, threatfox, urlhaus, virustotal) instead implement the new SourcePrecheck interface so the host UI can surface "not configured" before attempting a query.
Add two shared helpers to source.go and apply them across all sources:
- disabledResult(id, name) replaces the repeated inline SourceResult literal
- evidenceEval(r, severity) replaces the identical Evaluate body in 6 sources
Each source's Query() method previously set r.Listed and r.Severity,
embedding verdict logic inside the prober. Evaluation now lives in a
dedicated Evaluate(SourceResult) (bool, string) method per source,
keeping Query() as pure observation.
A package-level EvaluateResult() helper looks up the source by ID and
delegates to its Evaluate method; rules.go, report.go, types.go, and
provider.go all call this instead of reading pre-set r.Listed/r.Severity
values. An unknownSource sentinel handles results whose source is no
longer registered.
