Feature Request: Extend statusline JSON with detailed context breakdown and usage quota data

Problem

Currently, the statusline JSON input only provides aggregate token counts (input_tokens, cache_creation_input_tokens, cache_read_input_tokens). To monitor context and usage effectively, users must repeatedly run /context and /usage commands manually.

This is problematic because:

1. I often forget to check /usage until I've exhausted my quota mid-session
2. The detailed context breakdown from /context (system prompt, tools, MCP tools, agents, memory, messages) would help optimize context usage in real-time
3. There's no way to proactively monitor usage limits without interrupting workflow

Requested Additions to Statusline JSON

1. Detailed Context Breakdown (from /context)

{
  "context_window": {
    "breakdown": {
      "system_prompt_tokens": 3200,
      "system_tools_tokens": 16600,
      "mcp_tools_tokens": 1800,
      "custom_agents_tokens": 389,
      "memory_files_tokens": 837,
      "messages_tokens": 28100
    },
    "autocompact_buffer_size": 45000
  }
}

2. Usage Quota Data (from /usage)

{
  "usage_quota": {
    "session": {
      "used_percent": 20,
      "resets_at": "2024-12-18T20:00:00Z"
    },
    "weekly": {
      "used_percent": 44,
      "resets_at": "2024-12-22T12:00:00Z"
    },
    "weekly_sonnet": {
      "used_percent": 4,
      "resets_at": "2024-12-22T12:00:00Z"
    }
  }
}

Use Case

I built a custom statusline script (gist) that displays context usage visually. With the requested data, I could:

• Show a color-coded breakdown matching /context output
• Display usage quota warnings before hitting limits
• Make informed decisions about model switching (Opus vs Sonnet) based on remaining quota

Impact

This would significantly improve the Claude Code UX by enabling proactive resource monitoring without workflow interruption.