[EPIC] Remove Provider Health Monitoring System

[nickna]

Overview

Remove the entire provider health monitoring system from Conduit. Provider health checks will be handled externally via Python scripts or other monitoring solutions.

Motivation

• Current implementation is over-engineered for a simple monitoring task
• Timer-based background service has bugs and doesn't scale horizontally
• Adds unnecessary complexity to the codebase
• External monitoring provides more flexibility

Scope

• Remove all provider health monitoring code
• Fix all breaking dependencies
• Maintain core system health checks (database, Redis, RabbitMQ)
• Ensure no functionality regression

Success Criteria

• All provider health code removed
• System builds and tests pass
• WebUI gracefully handles missing health data
• Router continues to function without health-based filtering
• Metrics/monitoring adjusted for missing health data
• /health endpoint shows only core infrastructure status

Breaking Changes

• Provider health endpoints removed from Admin API
• WebUI provider health page removed
• Provider health metrics no longer exported
• Real-time health updates discontinued
• Health-based routing disabled

Migration Guide

External monitoring systems should:

1. Query provider list via Admin API
2. Implement their own health checks
3. Store results in their preferred monitoring system

Related Issues

Will be created as sub-issues of this epic.