Feature request: use tibbles, not vanilla data frames

Copying a short pitch I've made recently:

> Context: the tibble format is a popular data frame subclass used in many tidyverse packages like dplyr that tries to clean up some behavior of vanilla data.frames and slightly extend them.
> - (Short pitch: it has a nice print method that limits the number of rows printed so you don't get stuck waiting if you accidentally print without using head [and also limits the number of columns it prints data for, simply listing the others by name], changes the behavior of $ and [ to address some issues they can cause when writing functions, disables check.names and stringsAsFactors even pre-R-4.0, doesn't allow assigning row.names, allows some additional column types, etc.; its package also has the convenient tribble constructor.)

I think the print functionality would be pretty nice here.

No files, tests, or entry points are named. Start by locating where vanilla data frames are created and printed in the repository, then determine whether the request covers all data frames or only selected outputs. Done should include tibble-based behavior with bounded row and column printing, backed by tests for the affected outputs.

Written by the indexing model from the issue text.