feat(ai): introduce retrieval domain models and validation

[aotenjou]

Summary

Introduce the foundational retrieval domain model layer for the AI module.

This PR defines the typed data structures required by the retrieval pipeline, including source documents, text chunks, graph references, evidence, and version metadata. It also adds JSON serialization/deserialization and centralized model validation.

#864

Changes

• Add retrieval document models:
  • SourceDocument
  • SourceRef
  • TextChunk
• Add retrieval graph models:
  • EntityRef
  • GraphVertexRef
  • GraphEdgeRef
  • GraphPathRef
• Add retrieval evidence models:
  • Evidence
  • EvidenceKind
  • ChannelScore
• Add retrieval version models:
  • GraphVersion
  • IndexVersion
• Add JSON codec support through RetrievalModelJson
• Add centralized validation and RetrievalModelValidationException
• Use field-based equality semantics without introducing stable IDs
• Add unit tests and JSON fixtures covering model construction, equality, validation, and serialization behavior

How was this PR tested?

• Tests have Added for the changes
• Production environment verified

[aotenjou]

@yaozhongq @kitalkuyo-gita