Minor: improve tips display (#5631)

### What problem does this PR solve?

1. Add the missing translations.  
![CleanShot 2025-03-05 at 10 29
32](https://github.com/user-attachments/assets/85e95372-07d9-47a1-82cf-6eb4d0e1c831)

2. Shorten overly long tips.  
![CleanShot 2025-03-05 at 10 34
49](https://github.com/user-attachments/assets/fae8ce4c-6495-4abf-958d-2febeb38b893)

### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [x] Other (please describe):

web/src/components/parse-configuration/graph-rag-items.tsx

@@ -54,7 +54,7 @@ const GraphRagItems = ({ marginBottom = false }: GraphRagItemsProps) => {
     (title: React.ReactNode | string) => {
       return {
         title: typeof title === 'string' ? t(title) : title,
-        overlayInnerStyle: { width: '50vw' },
+        overlayInnerStyle: { width: '32vw' },
       };
     },
     [t],
@@ -67,7 +67,7 @@ const GraphRagItems = ({ marginBottom = false }: GraphRagItemsProps) => {
         label={t('useGraphRag')}
         initialValue={false}
         valuePropName="checked"
-        tooltip={renderWideTooltip('useGraphRagTip')}
+        tooltip={t('useGraphRagTip')}
       >
         <Switch />
       </Form.Item>

web/src/locales/en.ts

@@ -520,6 +520,7 @@ This auto-tag feature enhances retrieval by adding another layer of domain-speci
         'This optimizes user queries using context in a multi-round conversation. When enabled, it will consume additional LLM tokens.',
       howUseId: 'How to use chat ID?',
       description: 'Description of assistant',
+      descriptionPlaceholder: 'e.g. A chat assistant for resume.',
       useKnowledgeGraph: 'Use knowledge graph',
       useKnowledgeGraphTip:
         'It will retrieve descriptions of relevant entities,relations and community reports, which will enhance inference of multi-hop and complex question.',

web/src/locales/zh-traditional.ts

@@ -504,6 +504,8 @@ export default {
         '在多輪對話的中，對去知識庫查詢的問題進行最佳化。會呼叫大模型額外消耗token。',
       howUseId: '如何使用聊天ID？',
       description: '助理描述',
+      descriptionPlaceholder:
+        '例如 你是一個專業的簡曆助手，隻能回答簡曆的問題。',
       useKnowledgeGraph: '使用知識圖譜',
       useKnowledgeGraph提示:
         '它將檢索相關實體、關係和社區報告的描述，這將增強多跳和複雜問題的推理。',

web/src/locales/zh.ts

@@ -520,6 +520,8 @@ General：实体和关系提取提示来自 GitHub - microsoft/graphrag：基于
         '在多轮对话的中，对去知识库查询的问题进行优化。会调用大模型额外消耗token。',
       howUseId: '如何使用聊天ID？',
       description: '助理描述',
+      descriptionPlaceholder:
+        '例如 你是一个专业的简历助手，只能回答简历的问题。',
       useKnowledgeGraph: '使用知识图谱',
       useKnowledgeGraphTip:
         '它将检索相关实体、关系和社区报告的描述，这将增强多跳和复杂问题的推理。',