feat(agent): Add data analysis multi-agent application.​​

[Hikomorebi]

Description

This PR introduces a complete data analysis multi-agent framework for DB-GPT, including anomaly detection, volatility analysis, and automated report generation capabilities. The framework enables users to perform complex data analysis tasks through coordinated intelligent agents.

Key Features Added

1. Data Analysis Planner Agent: A specialized planner agent designed for data analysis tasks with customized constraints and examples.

2. Five Specialized Agents:

   • MetricInfoRetriever: Retrieves metric information from knowledge bases
   • AnomalyDetector: Detects anomalies by comparing baseline and current period data
   • VolatilityAnalyzer: Performs attribution analysis to identify root causes of metric anomalies
   • ReportGenerator: Integrates all analysis results into structured Markdown reports
   • DataScientist: Existing agent enhanced to support the new framework

3. Visualization Components: New visualization tags for anomaly detection, volatility analysis, and report generation.

4. Comprehensive Documentation: Detailed cookbook guide with step-by-step instructions, including:

   • Project deployment and environment setup
   • Dataset preparation with MySQL database schema
   • Knowledge base configuration
   • Application creation and usage instructions
   • Screenshots for visual guidance

Implementation Details

• Added new action classes for each specialized agent with proper input validation and error handling
• Implemented visualization protocols for consistent output formatting
• Enhanced the base agent to support report generation
• Created detailed documentation with practical examples using the Superstore dataset

How Has This Been Tested?

The framework has been tested with the Superstore dataset, demonstrating the complete workflow from data preparation to report generation. The multi-agent system successfully:

1. Retrieves metric information from knowledge bases
2. Detects anomalies in business metrics
3. Performs attribution analysis for identified anomalies
4. Generates comprehensive analysis reports

Manual testing was performed following the documented steps in the cookbook guide.

Snapshots

default.mp4

Checklist:

• My code follows the style guidelines of this project
• I have already rebased the commits and make the commit message conform to the project standard.
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• Any dependent changes have been merged and published in downstream modules

[Aries-ckt]

Thanks for your contribution!

[chenliang15405]

LGTM

[Aries-ckt]

LGTM