Rewrite StatisticPage with factual date-range analytics and KPI sections

[IanTheMitten]

Motivation

• Provide factual, non-sampled analytics driven by a standard date-range model (today, thisWeek, thisMonth, allTime, custom) so metric values reflect the selected boundaries.
• Centralize transaction filtering so every statistic is computed from the same filteredTransactions source and keep selectedPeriodId for period drill-downs.
• Surface high-level KPIs and factual sections (weekday, time-period, products, days, customers) for easier operational insight.

Description

• Replaced the sampling model and sampling UI with a simple date-range selector implemented on the page and a StatisticDateRange type including today, thisWeek, thisMonth, allTime, and custom.
• Derived filteredTransactions once with getDateBounds(...) and used it across all computed analytics including totalRevenue, unitsSold, transactionCount, avgBasketSize, weekday totals, time-period totals, top products, best days, and top customers.
• Added a KPI row rendering Total Revenue, Units Sold, Transactions, and Avg Basket Size.
• Reorganized the page into factual sections: weekday revenue chart, time-period revenue chart with clickable bars (keeps selectedPeriodId drill-down), selected-period running revenue chart, top products table, best days table, and top customers table.
• Removed sampling-related state/props and imports (samplingSeed, sampled modes/options, WeekdayRevenueBarChart, TimePeriodRevenueBarChart, TimePeriodCumulativeLine, ProductAverageRevenueTable, and analyticsSampling usages) and adopted inline chart/table rendering driven by filteredTransactions.

Testing

• Ran the production build with npm run build, which completed successfully (Vite build succeeded).

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4d7154c5c2

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".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Parse best-day keys as local dates

row.dayKey is produced in local calendar time (YYYY-MM-DD), but new Date(row.dayKey) parses that format as UTC; in US time zones this shifts many rows to the previous day (for example, 2026-04-01 can display as 3/31/2026). That makes the Best Days table show incorrect dates for the revenue/transaction totals it is summarizing, so users can act on the wrong day-level insight.

Useful? React with 👍 / 👎.