docs: add panel data example to frolladapt.Rd

[LeonidasZhak]

Summary

Adds a panel data example to showing how to compute rolling means within groups using + with by=. This is the data.table equivalent of Stata's rangestat with by() for time-aware rolling windows.

Stata equivalent:

rangestat (mean) rollmean3 = sales, int(date -2 0) by(firm)

data.table equivalent (new example):

dt = data.table(
  firm = rep(c("A", "B"), each = 4),
  date = as.Date(c("2020-01-01", "2020-01-02", "2020-01-05", "2020-01-06",
                    "2020-01-01", "2020-01-03", "2020-01-04", "2020-01-07")),
  sales = c(10, 12, 15, 13, 20, 22, 18, 25)
)
setorder(dt, firm, date)
dt[, rollmean3 := frollmean(sales, frolladapt(date, n=3L, partial=TRUE),
                            adaptive=TRUE), by = firm]

Rationale

The frolladapt.Rd page had only single-entity examples despite frolladapt being the key function for time-aware rolling windows in data.table. Panel data rolling operations (e.g., rangestat ... by(firm) in Stata) are the most common migration pattern where users need rolling calculations within groups respecting irregular time gaps.

The existing froll.Rd note (line 72) warns that rolling functions operate on physical order, but frolladapt.Rd didn't show the grouped usage pattern that makes this practical for panel data.

Validation

• tools::checkRd('man/frolladapt.Rd') passes (no output = no errors)
• All example code runs successfully in R
• 1 file changed, 13 lines added

Files Changed

File	Lines Added	Lines Removed	Type
man/frolladapt.Rd	13	0	Documentation (Rd)

Duplicate Check

• Checked open PRs: no frolladapt-related PRs found
• Checked issues: froll* could support classed n to detect adaptive automatically #7294 (about classed n for automatic adaptive detection, different scope)