Update lobe-chat to version v1.82.1 (#2612)

lobe-chat/docker-compose.yml

@@ -7,6 +7,6 @@ services:
       APP_PORT: 3210
 
   app:
-    image: lobehub/lobe-chat:1.79.10@sha256:76e79ba8a549c52927b15cd8aae815b9405c9cb139987625f2a7582ef0d850b0
+    image: lobehub/lobe-chat:1.82.1@sha256:4ad1f82d8c6d23d30f5887e225ac36e5ef0c9e387872eb7a7746d3d632dbcfc8
     restart: on-failure
     user: "1000:1000"

lobe-chat/umbrel-app.yml

@@ -3,7 +3,7 @@ id: lobe-chat
 name: Lobe Chat
 tagline: An open-source, modern-design AI chat framework
 category: ai
-version: "1.79.10"
+version: "1.82.1"
 port: 7455
 description: >-
   💬 An open-source, modern-design ChatGPT/LLMs UI/Framework.
@@ -30,13 +30,15 @@ defaultUsername: ""
 defaultPassword: ""
 dependencies: []
 releaseNotes: >-
-  This update includes improvements and style changes.
+  This update includes new features, improvements and bug fixes.
 
 
   Key highlights:
-    - Added GPT-4.1 model support
-    - Cleaned up TRPC structure for clearer responsibilities
-    - Made GPT-4.1-mini the default model
+    - Added support for Streamable HTTP MCP server and search1api as a search provider
+    - Introduced new model support: SenseNova-V6, SenseChat-Vision, 360 AI, Taichu, AI21, o3/o4-mini, step-r1, glm-z1
+    - Improved UI with style tweaks: token performance display, SystemRole folding, hotkeys, container borders
+    - Fixed bugs in model fetching, Gemini tool calling, search prompt, and NextAuth version lock
+    - Refactored code for user info fetching and feature documentation
 
 
   Full release notes are available at https://github.com/lobehub/lobe-chat/releases