fix(playwright): waitFor instead of waitForSelector for action popover button

[nmgarza5]

Description

For action popover, use locator.waitFor instead of page.waitForSelector

the latter is deprecated. we should make this switch across the playwright suite generally. my theory is the test runner is stalling and that issue is exacerbated by waitForSelector.

How Has This Been Tested?

existing

Additional Options

• [Optional] Override Linear Check

---

Summary by cubic

Removed networkidle wait in sendMessage and a redundant AI response check to reduce flakiness with streaming/background requests. Stabilized Action Management by waiting on locators before clicking, making tests faster and more reliable.

^{Written for commit c01f943. Summary will update on new commits.}

[cubic-dev-ai]

No issues found across 1 file

[greptile-apps]

Greptile Summary

Removed the waitForLoadState("networkidle") check from the sendMessage() E2E test utility function. The function already has explicit waits for the AI message to appear (30s timeout) and for the URL to update with chatId= (10s timeout), making the networkidle check redundant.

• Simplified test synchronization by removing potentially flaky networkidle wait
• Relies on more specific, explicit waits that directly verify expected behavior
• Change is isolated to test infrastructure and doesn't affect application code

Confidence Score: 4/5

• This PR is safe to merge with low risk
• The change removes a potentially unreliable wait condition while keeping two explicit, targeted waits that verify the actual expected behavior. The function already waits for the AI message to appear and the URL to update, making the networkidle check redundant. However, there's a slight concern about whether the networkidle wait was catching edge cases that the explicit waits might miss, though the existing waits appear comprehensive.
• No files require special attention

Important Files Changed

Filename	Overview
web/tests/e2e/utils/chatActions.ts	Removed waitForLoadState("networkidle") check after message sending - relies on existing explicit waits for AI message and URL change instead

Sequence Diagram

sequenceDiagram
    participant Test as E2E Test
    participant sendMessage as sendMessage()
    participant Page as Browser Page
    participant UI as Chat UI
    
    Test->>sendMessage: Call with message text
    sendMessage->>Page: Count existing AI messages
    Page-->>sendMessage: existingMessageCount
    sendMessage->>Page: Click chat input textarea
    sendMessage->>Page: Fill message text
    sendMessage->>Page: Click send button
    sendMessage->>Page: Wait for AI message count to increase
    Note over sendMessage,Page: Timeout: 30 seconds
    Page->>UI: Monitor for new AI message
    UI-->>Page: New AI message appears
    Page-->>sendMessage: Count increased
    sendMessage->>Page: Wait for URL to contain 'chatId='
    Note over sendMessage,Page: Timeout: 10 seconds
    Page-->>sendMessage: URL updated
    Note over sendMessage: REMOVED: waitForLoadState("networkidle")
    sendMessage-->>Test: Message sent successfully

Loading