[Docs] Add a guide on using AI to diagnose SeaTunnel issues from runtime logs

Background

Some users struggle to diagnose issues when encountering runtime errors. With the growing availability of AI tools (e.g., ChatGPT, Claude, Copilot, etc.), there is a great opportunity to help users leverage AI to analyze SeaTunnel runtime logs and quickly identify root causes — but no official guidance exists for this yet.

Proposal

We'd love to see a community-contributed PR that adds an official guide to the SeaTunnel documentation site covering:

• How to collect and prepare SeaTunnel runtime logs for AI analysis
• Recommended prompts / prompt templates for feeding logs into AI tools
• Common error patterns that AI can help identify (e.g., connector failures, checkpoint timeouts, OOM, task retries, etc.)
• Tips on filtering noise from logs before submitting to an AI
• Example walkthrough: from a real error log to a diagnosed root cause

Motivation

A well-written guide on this topic would significantly lower the barrier for new users to self-diagnose issues, reduce repetitive questions in the community, and demonstrate how AI tooling can complement SeaTunnel's observability story.

Call to Action

Is there anyone in the community willing to pick this up? A PR adding this guide to the official docs site would be hugely appreciated. Even a minimal first version covering basic log collection + prompt templates would be a great starting point.