RFile writes should utilize multiple threads

**Is your feature request related to a problem? Please describe.**
The write operations to an rfile are serialized. When writing large rfiles in map reduce jobs this can produces very large tales to the jobs. The bottleneck is often compression rather than i/o.

**Describe the solution you'd like**
Utilizing multiple threads to process multiple blocks in parallel could dramatically improve write performance. Having a dedicated thread to write completed blocks in order would still be necessary, but should be possible. This could be scaled based on available memory for buffering.

**Describe alternatives you've considered**
Adding pipelines to the existing code could be a smaller lift, and have a big performance improvement as well.

Start by locating the RFile write path and the code that compresses and writes blocks, then trace how map-reduce jobs serialize those operations. Compare the existing serialized path with the proposed parallel block processing and ordered output design. Done should include a measured write-performance improvement without changing block order or correctness.

Written by the indexing model from the issue text.