fix(client): card hover preview flicker with follow-cursor and delay

[kiannidev]

Summary

• When a deferred inspectObject(null) clear fires, keep the preview open if the cursor is still over a [data-card-hover] element
• Fixes follow-cursor + hover-delay settings where layout shifts caused brief preview dismissal

Fixes #4377

Test plan

• Added uiStore regression test for deferred clear suppression
• Enable follow-cursor preview with 400ms delay and confirm hover stays stable

Model: Composer

Made with Cursor

[gemini-code-assist]

Code Review

This pull request updates the uiStore to prevent the card preview from being cleared when the cursor is still over a card tile (using [data-card-hover]) during a spurious mouseleave event. A corresponding unit test has been added to verify this behavior. The review feedback recommends using Vitest's fake timers instead of a real setTimeout in the test to avoid potential flakiness in CI environments.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[matthewevans]

Maintainer sweep note: DEFER-FE. This PR touches frontend/client/UI surface, so I am leaving it for Matt's direct frontend handling under the sweep policy. No approval, label, enqueue, or merge from this automated maintainer sweep.

[matthewevans]

Maintainer sweep update: current head still touches frontend/client/UI surface, so this remains DEFER-FE for Matt's direct frontend handling. No approval, label, enqueue, or merge from this automated maintainer sweep.

[matthewevans]

Maintainer sweep update: current head 2a61f33e21320d3def1e8e45b630aae704c87efb still touches client UI store/tests. Per the frontend/client/UI sweep rule this remains DEFER-FE for Matt’s direct handling; no approval, label, enqueue, or merge from the sweep.

[matthewevans]

@kiannidev please show a recording illustrating that the issue has been reoslved

[kiannidev]

20260630_221334.mp4

[kiannidev]

@matthewevans - I uploaded video to show how it is working

[matthewevans]

Deferred by maintainer intake policy — not ignored.

This current head (a3564c77875dd97fe895bd2ec81a06afdca0fe04) was triaged as a frontend-only change (client/src/stores/__tests__/uiStore.test.ts, client/src/stores/uiStore.ts) by kiannidev. The local frontend-review allowlist does not include this author, so this route does not perform an implementation-diff review or approve the PR.

A maintainer must explicitly take this PR or add a local frontend-review exception before it can receive substantive review. The defer label is a routing marker only, not a verdict on the change.

[matthewevans]

Deferred by maintainer intake policy — not ignored.

This current head (249b46ab9cf87dfe787cdbaca031510df4157b76) was triaged as a frontend-only change (client/src/stores/__tests__/uiStore.test.ts, client/src/stores/uiStore.ts) by kiannidev. The local frontend-review allowlist does not include this author, so this route does not perform an implementation-diff review or approve the PR.

A maintainer must explicitly take this PR or add a local frontend-review exception before it can receive substantive review. The defer label is a routing marker only, not a verdict on the change.

[matthewevans]

Superseded by #6239.

Current main merged #6239 at 11238f4760a4be8038724596689827625e71f662), and it closes the same issue, #4377. Its uiStore.inspectObjectupdate covers this stale-leave case plus the pending hover-delay path: it preserves the pending show through the 50 ms clear grace period, flushes it when the pointer remains over[data-card-hover]or[data-card-preview], and cancels it otherwise. client/src/stores/tests/uiStoreHoverLatency.test.ts` exercises both the delayed and already-open preview cases.

I merged current main into this head locally to classify the conflict. The sole conflict is this same uiStore clear branch; keeping #6239's broader implementation leaves no unique behavior from this PR to port. Closing this duplicate rather than pushing a redundant merge.