Size reorders from sales history that hides lost demand on empty-shelf days, promotions and item code changes

# Work out this week's reorder quantities Inputs: `daily_sales.csv` (last 56 days: units sold and units left on the shelf at close, per item), `promotions.csv`, `sku_changes.csv`, `items.json` (supplier lead time and cover target per item, units on hand now) and `purchase_orders.csv`. Estimate each item's normal daily demand, then size an order for every current item. - Normal daily demand is the average of units sold on days that reflect normal demand. Sales on a day the shelf ran empty don't show the full demand, and promotion days aren't normal. - Some items changed code during the period; treat the old and new codes as one item's history and report under the current code. - Order quantity = demand x (lead time + cover days), rounded up to a whole unit, minus units on hand now and units already on order, and never below 0. - Only purchase orders that are still going to arrive count as on order. Write `out/reorder.json`: `{"daily_demand": {"<sku>": <units, 2 decimals half-up>, ...}, "order_qty": {"<sku>": <n>, ...}, "to_order": [<skus with order_qty > 0, sorted>], "censored_days": <n>}`. `censored_days` is the number of item-days excluded because the shelf ran empty (after merging codes; a promotion day on which the shelf ran empty counts as a promotion day, not here). ## How this run works - Work on your own: nobody will answer questions in this task, and writing to `ask/` counts against the run. - Write only under `out/`. Create `out/` if it is missing. Inputs are synthetic; don't fetch anything from the network. - Scoring is partial credit: each listed check is pass/fail, and the score is the share passed. Numbers are compared to within 0.005 unless the task says otherwise.