Add Analytics API service and Drush commands for AI integration

[jjroelofs]

Summary

Add a service-based analytics API to the RL module that enables both Drush commands and other Drupal modules (like ai_content_strategy) to access experiment performance data, insights, and recommendations.

Motivation

Currently, analyzing RL experiment data requires direct database queries. This makes it difficult for:

• AI tools (like Claude Code) to provide experiment insights via Drush
• Other Drupal modules to integrate with RL data
• Automation of experiment analysis and reporting

Proposed Solution

1. Service Layer (RlAnalyzer)

Create a service implementing RlAnalyzerInterface with methods:

• listExperiments() - All experiments with summary stats
• getStatus($experimentId) - Detailed experiment status
• getPerformance($experimentId, $options) - Arm performance with human-readable labels
• getTrends($experimentId, $options) - Historical data
• export($experimentId) - Full JSON export

2. Drush Commands

Thin CLI wrappers around the service:

• rl:list - List all experiments
• rl:status <experiment> - Experiment status
• rl:performance <experiment> - Arm performance
• rl:trends <experiment> - Historical trends
• rl:export <experiment> - Full data export

Key Features

• Human-readable labels - Resolve entity IDs to titles/names
• Pre-computed insights - vs_average, confidence levels, trends
• JSON output - Structured data for AI consumption
• Self-documenting - Rich help text for AI discoverability

Technical Approach

Following Drupal core patterns (like StateInterface/State):

1. Interface defines API contract
2. Service contains all business logic
3. Drush commands are thin wrappers
4. Other modules inject service via DI

Tasks

• Create RlAnalyzerInterface
• Create RlAnalyzer service
• Register in rl.services.yml
• Create RlCommands Drush class
• Implement all commands
• Update README.md
• Update docs/rl_project_desc.html