[niemyjski]

Usage charts stacked the allowance on top of usage, inflating the Limit line and allowing it to overflow the organization chart. Both routes now share independent series on a common baseline while preserving labels, colors, tooltip order, solid limit stroke, and project legend autoscaling. Adds seeded synthetic Storybook cases, component regressions, and a real-route browser regression. No API or billing-enforcement changes.

Validation: 4 component tests, type checks, formatting/ESLint, Storybook build, and local before/after comparisons passed. Hosted client, API, E2E, and Docker checks passed on a2e794f. Head 140d8c3 merges main 89607b4; only CONTRIBUTING.md changed, with application/tests unchanged. Final-head CI is running.

Verification and implementation details

• Compared the original main-branch routes with the fixed routes using the same seeded data in Chromium: 12 monthly hovers × both routes × light/dark themes. Tooltip text, row order, values, fonts, colors, legend styling, and line colors matched. Project Y-axis ticks change intentionally because series are no longer summed. Legend/footer screenshot pixels were identical.
• Regression tests cover constant/changing allowances, usage above the allowance, organization/project comparison totals, keyboard legend isolation/restoration, raw hover values, labels, and line containment. Tests and Storybook play functions use Arrange/Act/Assert sections; repository guidance now explicitly includes JavaScript/TypeScript.
• Storybook: Components/Organizations/UsageChart → Organization, Project, Empty, Changing Limits. Play tests retain full-series geometry assertions after exercising the legend.
• Fixtures contain generated counts and dates only. No customer names, identifiers, or copied history.
• Commands from src/Exceptionless.Web/ClientApp: npm run test:unit -- usage-chart.svelte.test.ts, npm run check, npm run lint, npm run build-storybook, and E2E_URL=<local Aspire App endpoint> npx playwright test e2e/tests/usage-charts.e2e.ts --project=chromium --workers=1 --retries=0.
• Initial aggregate validation hit local disk exhaustion. Type checks and lint subsequently passed individually. The later E2E failure occurred before chart navigation: login returned 500 with RedisTimeoutException, and Aspire reported the Docker socket unavailable. Hosted E2E subsequently passed on a2e794f, closing that verification gap. This task's AppHost was stopped after verification; no shared resources were reset.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-28T03:13:03.596439Z	a2e794f	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[github-actions]

Package	Line Rate	Branch Rate	Complexity	Health
Exceptionless.AppHost	23%	22%	130	❌
Exceptionless.Core	78%	70%	10988	✔
Exceptionless.Insulation	51%	43%	370	➖
Exceptionless.Web	86%	72%	9427	✔
Summary	81% (28620 / 35495)	70% (14328 / 20504)	20915	✔