[BUG] [Linux] Massive RAM spike (300MB → 10GB) due to background indexation

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

## Environment

• Claude Code version: 2.1.159
• OS: Ubuntu Linux, kernel 6.17.0-35-generic
• RAM: 16GB

## Description
Once or twice a week, when Claude Code starts, a background process progressively consumes RAM
from ~300MB up to ~10GB, then drops back to ~300MB once complete.

If the process is interrupted before completion (e.g. system crash due to OOM),
it restarts from scratch at the next launch.

## Project context

• Django + React project
• node_modules: 628MB (697 packages)
• A .claudeignore file excluding node_modules/, __pycache__/, logs/ etc.

was added but had no effect on RAM consumption.

## Impact
The RAM spike consistently crashes the system when other applications
(IDE, browser) are already using significant memory.

What Should Happen?

Indexation should either:

• Be cached persistently on disk so it doesn't repeat on every startup, or
• Use a reasonable amount of RAM (< 1GB) for a typical project, or
• Respect .claudeignore to reduce scope

Error Messages/Logs

Steps to Reproduce

1. Open Claude Code on a large Django/React project
2. Leave it idle without sending any prompt
3. Observe RAM usage climbing steadily from ~300MB to ~10GB over ~10 minutes
4. RAM drops back to ~300MB once the process completes

Claude Model

Sonnet (default)

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.159

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

PyCharm terminal

Additional Information

_No response_