Add governed-intelligence worked examples for claim and action flows

[mdheller]

Context

Follow-up from #77 after the governed-intelligence core ontology/schema/vector-manifest work landed.

Current main includes the canonical governed-intelligence model and related contract surfaces, including:

• Platform/GovernedIntelligence/governed-intelligence.ttl
• shapes/governed_intelligence.shacl.ttl
• contexts/governed-intelligence.context.jsonld
• catalog/governed_intelligence_registry.ttl
• docs/specs/governed-intelligence-object-model-v0.md
• schemas/governed-intelligence.v1.schema.json
• manifests/governed-intelligence.vector-encoding.manifest.v1.json
• bindings/governed_intelligence/README.md

The remaining acceptance criteria from #77 are the worked example flows.

Scope

Add governed-intelligence worked examples for:

1. Text evidence span into Anchor -> Evidence -> Claim -> PolicyDecision.
2. Agent action into ActionProposal -> ActionAdmission -> RuntimeReceipt.
3. Vector candidate example showing candidate_only status and non-admitted claim boundary.

Required files

Suggested locations:

• examples/governed-intelligence/evidence-policy-demo.ttl
• examples/governed-intelligence/action-receipt-demo.ttl
• examples/governed-intelligence/vector-candidate-demo.ttl

Acceptance criteria

• Examples use the canonical Platform governed-intelligence namespace.
• Examples conform to shapes/governed_intelligence.shacl.ttl.
• Examples are discoverable from docs/module-map.md or an examples README.
• make validate / CI remains green.

Non-goals

• No Holmes reasoning runtime.
• No Sherlock retrieval runtime.
• No GAIA fusion runtime.
• No Agentplane execution implementation.