Add arrow feature to re_chunk and conversions to RecordBatch

[jleibs]

What

Basic type conversions from TransportChunk to RecordBatch and back.

Adding the round-trip test turned up an interesting issue.

TransportChunk <-> RecordBatch fails to round-trip successfully because we lose the ExtensionType encapsulation that used to be encoded by arrow2.

While on the surface this isn't immediately problematic, as we don't care about ExtensionTypes, the discussion indicates there are in fact going to be very real pain points when it comes to writing semantic data processing engines using arrow-rs. This is because the metadata is attached to the FIELD, not the DATATYPE, and there exist many processing contexts where the context of that field itself is lost.
apache/arrow-rs#4472

Checklist

• I have read and agree to Contributor Guide and the Code of Conduct
• I've included a screenshot or gif (if applicable)
• I have tested the web demo (if applicable):
  • Using examples from latest main build: rerun.io/viewer
  • Using full set of examples from nightly build: rerun.io/viewer
• The PR title and labels are set such as to maximize their usefulness for the next release's CHANGELOG
• If applicable, add a new check to the release checklist!
• If have noted any breaking changes to the log API in CHANGELOG.md and the migration guide

• PR Build Summary
• Recent benchmark results
• Wasm size tracking

To run all checks from main, comment on the PR with @rerun-bot full-check.

[teh-cmc]

very nice

[teh-cmc]

Hate to be that guy but im gonna be that guy

Suggested change

	pub fn try_as_arrow_record_batch(&self) -> Result<RecordBatch, ArrowError> {
	pub fn try_to_arrow_record_batch(&self) -> Result<RecordBatch, ArrowError> {