feat: implement fuzzy search and replace functionality with Levenshtein distance

- Added `applySearchReplace` function to handle search and replace operations with fuzzy matching capabilities.
- Introduced tests for various scenarios including fuzzy matching with typos, exact matches, and handling whitespace differences.
- Created a parser for search/replace blocks to facilitate the new functionality.
- Updated prompts for search-replace operations to clarify usage and examples.
- Added utility functions for text normalization and language detection based on file extensions.
- Implemented a minimal stdio MCP server for local testing with tools for adding numbers and printing environment variables.

testing/fake-llm-server/README.md

@@ -7,6 +7,7 @@ A simple server that mimics the OpenAI streaming chat completions API for testin
 - Implements a basic version of the OpenAI chat completions API
 - Supports both streaming and non-streaming responses
 - Always responds with "hello world" message
+- Simulates a 429 rate limit error when the last message is "[429]"
 - Configurable through environment variables
 
 ## Installation
@@ -82,6 +83,29 @@ For non-streaming requests, you'll get a standard JSON response:
 
 For streaming requests, you'll receive a series of server-sent events (SSE), each containing a chunk of the response.
 
+### Simulating Rate Limit Errors
+
+To test how your application handles rate limiting, send a message with content exactly equal to `[429]`:
+
+```json
+{
+  "messages": [{ "role": "user", "content": "[429]" }],
+  "model": "any-model"
+}
+```
+
+This will return a 429 status code with the following response:
+
+```json
+{
+  "error": {
+    "message": "Too many requests. Please try again later.",
+    "type": "rate_limit_error",
+    "param": null,
+    "code": "rate_limit_exceeded"
+  }
+}
+```
 
 ## Configuration