Improve performance of RunArray --> Row conversion

[alamb]

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

• While reviewing arrow-row: Add support for REE #7649 I had some thoughts about how to improve performance of RunArray --> Row conversion

Describe the solution you'd like

1. Add a RunArray to the existing row benchmarks: https://github.com/apache/arrow-rs/blob/e5fb84a482162cf5ba40be05af72e74c47898381/arrow/benches/row_format.rs#L75-L74
2. Potentially try and optimize the performance, using the benchmark to guide the changes

Describe alternatives you've considered

1. You could hoist this out of the inner loop so it was executed once per physical value rather than once per logical value
2. You could potentially encode row once and then simply copy the encoded bytes for all remaining rows. This is probably significantly faster than re-encoding the same value over and over again.

Originally posted by @alamb in #7649 (comment)

Additional context

[sonhmai]

take