docs: Broader introduction - problem, solution, strengths

[ijac13]

Summary

Restructure the landing page to better communicate what Recce is and why it matters.

Changes

1. Problem/Value (why Recce exists)

• Data pipelines succeed but data is quietly wrong
• PRs merge without checking what actually changed in the data
• Validation knowledge stays in people's heads instead of becoming team practice
• Junior and senior engineers apply different standards

2. Solution (what Recce is)

Core: Validation engine that compares base vs. current environments and visualizes differences.

Access via:

Method	Description
Cloud	Hosted service with automated PR validation
OSS	Self-hosted validation engine
CLI	Command-line interface for local and CI/CD
Agent	AI-powered reviewer that analyzes changes
Plugin	Claude Code plugin for guided setup
MCP	Model Context Protocol for AI agents

3. What Makes Recce Different (strengths)

• Column-level impact radius: Validate only affected columns, not entire models. Key differentiator.
• Agent as reviewer zero: Agent does validation, surfaces unknown unknowns. Users focus on judgment.
• Collaborate and standardize: Individual checks → team standards (preset checks)

Test Plan

• mkdocs build succeeds
• Review content accuracy
• Verify links work

---

🤖 Generated with Claude Code