Consider making the DataFrame API less verbose by imitating Polars

**Is your feature request related to a problem or challenge? Please describe what you are trying to do.**

I'd say that, for instance, `.select(["name"])` is significantly easier to read and write than `.select(vec![col("name")])`.

**Describe the solution you'd like**

Polars deals with this by having the following more generic signature.

```
pub fn select(&self, selection: I) -> Result
where
I: IntoIterator,
S: AsRef
```

`IntoIterator` makes a lot of sense, it allows for arrays for concise syntax and Vec or any other collection whenever the list is programmatically generated.

`AsRef` is less clear, I'm not sure how they refer to complex expressions with a string reference. Perhaps `Into` might be better with a default implementation for `&str`.

Yes I know that there's also a `select_columns(&self, columns: &[&str])` method in the current API, but it is still less flexible, somewhat redundant, and there are other places where `vec!` and `col()` somewhat pollute the syntax still.

Start by locating the DataFrame API methods named select and select_columns, then inspect their current signatures and nearby usage. Compare the proposed generic selection behavior with existing column and complex-expression handling; done should mean an agreed, consistent API design that covers concise literals and programmatically generated collections without leaving expression behavior unresolved.

Written by the indexing model from the issue text.