[DataFrame] Read files in parallel (4x faster)

[alamb]

Which issue does this PR close?

Closes #6983
Closes #6908

Rationale for this change

This code uses a single core to read the file

    let _df = _ctx.read_parquet(FILENAME, _read_options).await.unwrap();
    let _cached = _df.cache().await;

What changes are included in this PR?

Use multiple cores

Testing using using cargo --release

With main (16s)

datafusion end -> 2023-07-16T09:07:29.895269-04:00 16.080133858s

With this branch (3s)

datafusion end -> 2023-07-16T08:52:05.511984-04:00 2.947019517s

Are these changes tested?

Yes

Are there any user-facing changes?

[alamb]

I think a better approach is described on #6983 (comment)