fix(knowledge): prevent abnormal deletion of local SQL files during knowledge base removal

[GeorgeDong32]

What this PR does

Before this PR:

删除知识库时，对应的本地数据库可能会因为来不及释放文件占用而导致删除失败

After this PR:

删除前释放RagApplication占用，添加删除异常重试机制实现文件的正确删除

Fixes #7592

Why we need it and why it was done in this way

The following tradeoffs were made:

The following alternatives were considered:

Links to places where the discussion took place:

Breaking changes

If this PR introduces breaking changes, please describe the changes and the impact on users.

Special notes for your reviewer

Checklist

This checklist is not enforcing, but it's a reminder of items that could be relevant to every PR.
Approvers are expected to review this list.

• PR: The PR description is expressive enough and will help future contributors
• Code: Write code that humans can understand and Keep it simple
• Refactor: You have left the code cleaner than you found it (Boy Scout Rule)
• Upgrade: Impact of this change on upgrade flows was considered and addressed if required
• Documentation: A user-guide update was considered and is present (link) or not required. You want a user-guide update if it's a user facing feature.

Release note

[Copilot]

Pull Request Overview

Provides robust cleanup when removing a knowledge base by closing database connections and retrying file deletions to avoid “file busy” errors.

• Track and reuse RAGApplication and LibSqlDb instances per knowledge base
• Close database client before removal and pause to ensure file handles are released
• Implement a retry loop with backoff for deleting the local storage directory

Comments suppressed due to low confidence (2)

src/main/services/KnowledgeService.ts:104

• [nitpick] Comments are in Chinese; consider translating to English to maintain consistency with the rest of the codebase.

  // 存储数据库实例，用于后续关闭

src/main/services/KnowledgeService.ts:205

• The new retry logic isn’t covered by existing tests. Add unit or integration tests to simulate busy-file scenarios and verify correct retry behavior.

      for (let i = 0; i < maxRetries; i++) {

[alephpiece]

网络搜索那边也用了 delete，我发现有些情况下会一直重试：

• 开启网络搜索 RAG，先发送一条用户消息 A
• 然后修改网络搜索 RAG 的模型
• 找到用户消息 A，重新发送

是我的用法有问题吗？我给网络搜索 RAG 建立了临时知识库，在配置变化的时候会先 delete，然后重新创建

[alephpiece]

cherry-studio/src/renderer/src/services/WebSearchService.ts

Lines 222 to 230 in 71917eb

	// 如果已存在且配置未变，直接复用
	if (state.searchBase && this.isConfigMatched(state.searchBase, config)) {
	return state.searchBase
	}
	// 清理旧的知识库
	if (state.searchBase) {
	await window.api.knowledgeBase.delete(state.searchBase.id)
	}

[GeorgeDong32]

网络搜索那边也用了 delete，我发现有些情况下会一直重试：

这个我还真没注意，我明天研究一下

[YinsenHo]

好