v3.18 (#3800)

* v3.18

* Update webview-ui/src/i18n/locales/ko/chat.json

Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>

---------

Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>

Author: mrubens

.changeset/v3.18.0.md

@@ -0,0 +1,23 @@
+---
+"roo-cline": minor
+---
+
+- Add support for Gemini 2.5 Flash preview models (thanks @shariqriazz and @daniel-lxs!)
+- Add button to task header to intelligently condense content with visual feedback
+- Add YAML support for mode definitions (thanks @R-omk!)
+- Add allowedMaxRequests feature to cap consecutive auto-approved requests (inspired by Cline, thanks @hassoncs!)
+- Add Qwen3 model series to the Chutes provider (thanks @zeozeozeo!)
+- Fix more causes of grey screen issues (thanks @xyOz-dev!)
+- Add LM Studio reasoning support (thanks @avtc!)
+- Add refresh models button for Unbound provider (thanks @pugazhendhi-m!)
+- Add template variables for version numbers in announcement strings (thanks @ChuKhaLi!)
+- Make prompt input textareas resizable again
+- Fix diffview scroll display (thanks @qdaxb!)
+- Fix LM Studio and Ollama usage tracking (thanks @xyOz-dev!)
+- Fix links to filename:0 (thanks @RSO!)
+- Fix missing or inconsistent syntax highlighting across UI components (thanks @KJ7LNW!)
+- Fix packaging to include correct tiktoken.wasm (thanks @vagadiya!)
+- Fix import settings bugs and position error messages correctly (thanks @ChuKhaLi!)
+- Move audio playing to the webview to ensure cross-platform support (thanks @SmartManoj and @samhvw8!)
+- Simplify loop syntax in multiple components (thanks @noritaka1166!)
+- Auto reload extension core changes in dev mode (thanks @hassoncs!)

README.md

@@ -49,13 +49,13 @@ Check out the [CHANGELOG](CHANGELOG.md) for detailed updates and fixes.
 
 ---
 
-## 🎉 Roo Code 3.17 Released
+## 🎉 Roo Code 3.18 Released
 
-Roo Code 3.17 brings powerful new features and improvements based on your feedback!
+Roo Code 3.18 brings powerful new features and improvements based on your feedback!
 
-- **Implicit Caching for Gemini** - Gemini API calls are now automatically cached, reducing API costs.
-- **Smarter Mode Selection** - Mode definitions can now include guidance on when each mode should be used, enabling better orchestration.
-- **Intelligent Context Condensing** - Intelligently summarizes conversation history when context fills up instead of truncating (enable in Settings -> Experimental).
+- **Gemini 2.5 Flash Preview Models** - Access the latest Gemini Flash models for faster and more efficient responses.
+- **Intelligent Context Condensing Button** - New button in task header lets you intelligently condense content with visual feedback.
+- **YAML Support for Mode Definitions** - Create and customize modes more easily with YAML support.
 
 ---