Optimise load config in UI

[lifeizhou-ap]

Summary

1. Optimise the load time when app starts or new session starts

   • Reduce the call to check ELEVENLAB_API_KEY exists or not. This call is expensive as takes around 3s because it checks keychain at the backend

     • Consolidate dictation settings retrieval to eliminate duplicate hook calls which makes the duplicated api call
     • Only check ELEVENLAB_API_KEY when the provider is eleven_lab, otherwise no need to check
     • Add module-level caching for ElevenLabs API key checks to avoid redundant keychain reads

   • Fix getProviders callback memoization in ConfigContext using a ref to prevent unnecessary recreations

2. Enhanced user experience to show the puzzle of current session extension only after the extensions are loaded.

Type of Change

• Feature
• Bug fix
• Refactor / Code quality
• Performance improvement
• Documentation
• Tests
• Security fix
• Build / Release
• Other (specify below)

AI Assistance

• This PR was created or reviewed with AI assistance

Testing

Manual testing by adding sleep in the /config/read endpoint

Before
https://github.com/user-attachments/assets/2818649f-fce3-48de-806f-861de3612d74

After
https://github.com/user-attachments/assets/56546fdc-c722-4d5a-a7b5-5adde2f66b19

[Copilot]

Pull request overview

This PR optimizes config loading in the desktop UI to improve performance, particularly around dictation settings and provider configuration. The main issue being addressed appears to be excessive calls to read config values (especially secret keys from the keychain).

Changes:

• Add module-level caching for ElevenLabs API key checks to avoid redundant keychain reads
• Fix getProviders callback memoization in ConfigContext using a ref to prevent unnecessary recreations
• Consolidate dictation settings retrieval to eliminate duplicate hook calls

Reviewed changes

Copilot reviewed 11 out of 11 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
ui/desktop/src/hooks/useWhisper.ts	Add constant for elevenlabs provider and export dictationSettings
ui/desktop/src/hooks/useDictationSettings.ts	Implement module-level cache for ElevenLabs key with conditional loading
ui/desktop/src/hooks/dictationConstants.ts	Add DICTATION_PROVIDER_ELEVENLABS constant and isSecretKeyConfigured helper
ui/desktop/src/components/settings/dictation/VoiceDictationToggle.tsx	Replace string literal with constant
ui/desktop/src/components/settings/dictation/ProviderSelector.tsx	Replace string literals with constant
ui/desktop/src/components/settings/dictation/ProviderInfo.tsx	Replace string literal with constant
ui/desktop/src/components/settings/dictation/ElevenLabsKeyInput.tsx	Use helper function and update cache, remove debug logs
ui/desktop/src/components/bottom_menu/BottomMenuExtensionSelection.tsx	Add loading state to prevent button visibility flicker
ui/desktop/src/components/ConfigContext.tsx	Fix getProviders memoization with ref pattern
ui/desktop/src/components/ChatInput.tsx	Get dictationSettings from useWhisper, remove duplicate hook call
crates/goose-server/src/routes/config_management.rs	Add 3-second debug sleep for secret config reads (TODO comment)

[Copilot]

This debug code adds a 3-second delay to all secret config reads and should not be merged to production. Remove this entire block before merging.

Suggested change

	// TODO: Remove this - simulating slow keychain access for debugging
	if query.is_secret {
	tokio::time::sleep(std::time::Duration::from_secs(3)).await;
	}

[zanesq]

Thanks!