Define default behavior for flattening results of `unfold` and `groupBy` aggregations

### Problem

Results of `unfold` and `groupBy` aggregations often produce `FrameColumn` (nested structure), which users typically have to manually flatten.

This leads to repetitive patterns:

df.groupBy { ... }.aggregate { ... }.flatten()

In many data processing workflows, users expect a flat tabular result by default.

### Design question

Should results of `unfold` and aggregation operations be flattened automatically?

Options:

- keep current behavior (manual `flatten()`)
- auto-flatten results by default
- introduce an explicit option (e.g. `flatten = true`)

### Expected

Define a consistent default behavior for result shape of these operations.

### Acceptance criteria

- Decision made on default behavior (manual vs automatic flattening)
- Behavior implemented or explicitly documented
- If auto-flatten is introduced:
- nested structures are flattened consistently
- If not:
- documentation explains when `flatten()` is required
- Add example covering aggregation + flattening

### Motivation

DataFrame is designed for tabular data processing, where flat structures are the common case :contentReference[oaicite:0]{index=0}

Requiring manual flattening in common workflows reduces usability and leads to boilerplate.

This behavior must be clearly defined before 1.0, as it affects the shape of core API results.

```
When using `unfold` or `groupBy` aggregations, it always feels bit cumbersome to call `flatten` on the result to pull the columns out of the FrameColumn.

I'd argue that by default the user always expects a flattened result because that's the behavior in r and python.
```