[FIX] pivot: carry forward running totals for missing date buckets

[dhrp-odoo]

Description:

When a measure is displayed as "running total" and the pivot is grouped by a date granularity (month/day/etc), PIVOT.VALUE returned an empty value for a requested bucket that does not exist in the pivot. This happened because the running-total cache only stores values for materialized pivot buckets, so a missing date key had no direct lookup.

With this commit, for date/datetime dimensions, we build an ordered index of existing buckets per secondary domain and running-total group, and use a binary search to find the nearest previous bucket. If found, we return its cumulative value (carry-forward); if the requested bucket is before the first one, we return empty.

Task: 5420999

review checklist

• feature is organized in plugin, or UI components
• support of duplicate sheet (deep copy)
• in model/core: ranges are Range object, and can be adapted (adaptRanges)
• in model/UI: ranges are strings (to show the user)
• undo-able commands (uses this.history.update)
• multiuser-able commands (has inverse commands and transformations where needed)
• new/updated/removed commands are documented
• exportable in excel
• translations (_t("qmsdf %s", abc))
• unit tested
• clean commented code
• track breaking changes
• doc is rebuild (npm run doc)
• status is correct in Odoo

[robodoo]