Design and implement route history tracking

[HyeockJinKim]

Story Description

Extend the history tracking system to include routing operations, tracking traffic routing decisions and load balancing history.

Acceptance Criteria

• Design route history schema and models
• Create RouteHistoryRepository in repositories layer
• Integrate with existing routing/load balancing logic
• Track route selection decisions
• Record traffic distribution patterns
• Monitor route health and performance metrics
• Track route failures and recovery
• Implement query methods for route history
• Add comprehensive testing

Schema Design

Main Table: route_history

- id: UUID (Primary Key)
- route_id: UUID
- session_id: UUID (Optional, Foreign Key to sessions)
- endpoint_id: UUID (Optional)
- route_type: Enum (session/service/inference)
- source_info: JSONB (client IP, user agent, etc.)
- target_info: JSONB (selected backend, agent, etc.)
- routing_decision: JSONB (algorithm used, weights, etc.)
- status: Enum (routed/failed/timeout/rejected)
- latency_ms: Integer
- error_code: String (Optional)
- message: Text
- created_at: DateTime

Sub-table: route_history_metrics

- id: UUID (Primary Key)
- history_id: UUID (Foreign Key to route_history)
- metric_name: String
- metric_value: Float
- metric_unit: String
- timestamp: DateTime

Repository Methods

async def record_route_decision(...)
async def update_route_status(...)
async def add_route_metrics(...)
async def get_route_history(session_id)
async def get_route_statistics(time_range)
async def get_failed_routes(time_range)

Integration Points

• App proxy routing decisions
• Load balancer selections
• Service endpoint routing
• Inference endpoint routing
• WebSocket connection routing

Metrics to Track

• Route selection time
• Backend response time
• Total request latency
• Route failure rate
• Traffic distribution patterns

JIRA Issue: BA-2443