[automatic failover] Implement wait on healthCheck results during client init

[atakavci]

This PR is based on changes in previous #4204.

Summary of changes in PR;

• Added initialization synchronization with health monitoring - StatusTracker tracks initial health results and manages event-driven waits for health checks and prevents event processing during startup
• Improved health check registration and fallback - registers listeners before clusters, defaults to HEALTHY when no strategy
• Enhanced failback mechanism for better visibility - made periodicFailbackCheck testable and improved thread naming
• Added forced active cluster with duration control - allows temporary cluster forcing with time limits
• Enhanced test utilities and helper methods - centralized health status change triggering and better test support
• Added graceful degradation with proper error handling - throws exceptions when all clusters unhealthy, better timeout messages
• Downgraded logback version - for compatibility reasons with slf4j version

Commits essential to this one are;

• - introduce StatusTracker with purpose of waiting initial healthcheck…

• - introduce forceActiveCluster by duration

• - fix failing tests by waiting on clusters to get healthy

• - fix failing scenario test

• - adressing reviews and feedback

[Copilot]

Pull Request Overview

This PR implements automatic failover with wait on health check results during client initialization. It introduces status tracking to wait for initial health check results and adds significant new functionality for health monitoring and automatic cluster management.

• Adds comprehensive health check system with StatusTracker for waiting on initial health check results
• Implements weighted cluster selection and automatic failback mechanisms with grace periods
• Replaces index-based cluster management with endpoint-based approach for better flexibility

Reviewed Changes

Copilot reviewed 33 out of 33 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
MultiClusterPooledConnectionProvider.java	Core provider changes - adds health status management, weight-based selection, and failback scheduling
StatusTracker.java	New class for event-driven health status waiting during initialization
HealthStatusManager.java	New health status management system with listener support
HealthCheck.java	Health check execution with timeout and callback mechanisms
EchoStrategy.java	Default health check strategy using Redis ECHO command
MultiClusterClientConfig.java	Configuration updates for weights, health checks, and failback settings
Various test files	Comprehensive test coverage for new health check and failback functionality

Comments suppressed due to low confidence (1)

src/test/java/redis/clients/jedis/mcf/StatusTrackerTest.java:178

• The test expects a JedisConnectionException but doesn't verify the specific exception type or message. Should use assertThrows() for better exception verification.

                fail("Should have thrown JedisConnectionException due to interrupt");