Reduce dependencies of datafusion-sql crate

[mbrobbel]

Which issue does this PR close?

This doesn't close, but contributes to the changes being worked on in #1750.

Rationale for this change

By disabling features and removing unused dependencies the build time of those crates is reduced.

What changes are included in this PR?

This PR removes some unused direct dependencies of the sql, common and expr crates, and disables unused features.

Are there any user-facing changes?

No

[codecov-commenter]

Codecov Report

Merging #3566 (f522f88) into master (81b5794) will increase coverage by 0.10%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #3566      +/-   ##
==========================================
+ Coverage   85.81%   85.91%   +0.10%     
==========================================
  Files         300      301       +1     
  Lines       55548    56218     +670     
==========================================
+ Hits        47666    48301     +635     
- Misses       7882     7917      +35     

Impacted Files	Coverage Δ
datafusion/sql/src/planner.rs	81.25% <100.00%> (ø)
datafusion/optimizer/src/type_coercion.rs	92.18% <0.00%> (-1.38%)	⬇️
...sion/core/src/physical_plan/file_format/parquet.rs	94.65% <0.00%> (-0.25%)	⬇️
datafusion/expr/src/logical_plan/plan.rs	77.75% <0.00%> (-0.17%)	⬇️
datafusion/core/tests/sql/predicates.rs	100.00% <0.00%> (ø)
...tafusion/core/src/physical_plan/file_format/mod.rs	96.94% <0.00%> (ø)
datafusion/optimizer/src/reduce_cross_join.rs	94.54% <0.00%> (ø)
datafusion/core/src/execution/context.rs	79.33% <0.00%> (+0.02%)	⬆️
datafusion/optimizer/src/projection_push_down.rs	98.11% <0.00%> (+0.04%)	⬆️
datafusion/common/src/scalar.rs	85.18% <0.00%> (+0.06%)	⬆️
... and 6 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[andygrove]

LGTM. Thanks @mbrobbel

[tustvold]

Given the only reason for these dependencies is deriving From conversions for ArrowError, perhaps we could/should just remove them. This is what I have been doing when splitting apart arrow-rs - see apache/arrow-rs#2711.

In general the error handling is a little bit unfortunate, so perhaps this will all get addressed eventually by apache/arrow-rs#2725

[mbrobbel]

Yes absolutely. I was planning on revisiting this after the arrow-rs split.

[tustvold]

Just as an FYI, mixing feature combinations within a workspace can make for a pretty poor development experience as switching between targets causes things to get rebuilt. Hopefully as we split apart arrow-rs apache/arrow-rs#2594 our reliance on feature flags will reduce.

See https://docs.rs/cargo-hakari/latest/cargo_hakari/about/index.html#what-are-workspace-hack-crates for more information.

[alamb]

Thanks @mbrobbel

[ursabot]

Benchmark runs are scheduled for baseline = cec6395 and contender = 31acc83. 31acc83 is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ec2-t3-xlarge-us-east-2] ec2-t3-xlarge-us-east-2
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on test-mac-arm] test-mac-arm
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ursa-i9-9960x] ursa-i9-9960x
[Skipped ⚠️ Benchmarking of arrow-datafusion-commits is not supported on ursa-thinkcentre-m75q] ursa-thinkcentre-m75q
Buildkite builds:
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python, R. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java