[diegosouzapw]

F5.1 coverage gaps — compression track

The F5.1 traceability audit (item→code→test) of the compression track (rodada4 master
plan, ~19 items) found the track ~16/19 done+tested, with three untested units —
code present and correct, but no covering test. This PR closes them (tests only).

Gap	Unit	What was missing
F3.2	useCompressionReplay reducer + stepMs	Only buildReplayFrames was tested; the replay state machine (play / pause / tick / auto-stop-at-last-frame / reset / set-speed) and the speed→interval pacing had zero assertions.
F3.3	useLiveCompression → accumulateRun	Exported "for unit tests" but never tested. Now pins most-recent-first ordering, the maxRuns cap (drops oldest), and input-array immutability.
F0.2	StatusDot (U0)	No dedicated render test. Pins the error-color override and sizeClass via createRoot+act (jsdom).

Scope

Tests only. The single production change is visibility-only: replayReducer,
INITIAL_STATE, stepMs, and the ReplayState/ReplayAction types are exported from
useCompressionReplay.ts as unit-test seams (no logic change), matching the codebase's
pattern of exposing pure functions for direct testing (cf. compressionFlowModel).

Tests

17 new tests, all green: 14 node:test (compression-replay-reducer,
live-compression-accumulate) + 3 vitest (status-dot). typecheck:core / lint /
check:cycles clean.

Context

Part of the F5 consolidation pass. Remaining compression-track items tracked separately:
F2.4 benchmark A/B (VPS-gated), F3.3 real mid-pipeline compression.step streaming
(scope decision — current impl is a client-side replay of one compression.completed
event), and F4.4 quantization (PRs #4187 + #4190).

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[gemini-code-assist]

Code Review

This pull request exports internal state machine elements from useCompressionReplay.ts to serve as unit-test seams and introduces new unit tests for the compression replay reducer, the live compression run accumulator, and the StatusDot UI component. The review feedback suggests tracking and properly unmounting React 18 roots inside act during test cleanup to prevent memory leaks and ensure lifecycle effects are flushed.

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.

[gemini-code-assist]

In React 18, when using createRoot, failing to call root.unmount() can lead to memory leaks and unexpected side effects (such as cleanup effects running after the DOM container has been removed). We should track the created roots and unmount them during cleanup.

Suggested change

	const containers: HTMLElement[] = [];
	function mount(ui: React.ReactElement): HTMLElement {
	const container = document.createElement("div");
	document.body.appendChild(container);
	containers.push(container);
	const root = createRoot(container);
	act(() => {
	root.render(ui);
	});
	return container;
	}
	const roots: { root: ReturnType<typeof createRoot>; container: HTMLElement }[] = [];
	function mount(ui: React.ReactElement): HTMLElement {
	const container = document.createElement("div");
	document.body.appendChild(container);
	const root = createRoot(container);
	act(() => {
	root.render(ui);
	});
	roots.push({ root, container });
	return container;
	}

[gemini-code-assist]

Unmount the React roots inside act to ensure all React lifecycle cleanups are properly flushed and to prevent memory leaks.

afterEach(() => {
  act(() => {
    while (roots.length > 0) {
      const item = roots.pop();
      if (item) {
        item.root.unmount();
        item.container.remove();
      }
    }
  });
  document.body.innerHTML = "";
});