Allow sorting by aggregated groups

[isidentical]

Which issue does this PR close?

Closes #2360.

Rationale for this change

It seems like we used to only collect column references from the actual aggregate expressions, not the grouped expressions. This PR changes the processing of rewrite_sort_col_by_aggs to also include that fact.

What changes are included in this PR?

Expression rewriter inside rewrite_sort_col_by_aggs now considers both distinct grouping expressions (e.g. GROUP BY <something>) as well as regular aggregate expressions when looking for a rewrite.

This PR also includes another change which corrects the behaviour of add_missing_columns when it is re-building the Projection. We used to ignore applying the alias to the field qualifiers in the schema, but with this patch that is also fixed (needed by the test sql::explain_analyze::explain_analyze_baseline_metrics).

Are there any user-facing changes?

This is a bug fix.

[codecov-commenter]

Codecov Report

Merging #3280 (6e6b674) into master (eedc787) will increase coverage by 0.07%.
The diff coverage is 96.87%.

@@            Coverage Diff             @@
##           master    #3280      +/-   ##
==========================================
+ Coverage   85.84%   85.92%   +0.07%     
==========================================
  Files         291      294       +3     
  Lines       52981    53470     +489     
==========================================
+ Hits        45484    45946     +462     
- Misses       7497     7524      +27     

Impacted Files	Coverage Δ
datafusion/expr/src/expr_rewriter.rs	86.26% <80.00%> (+0.17%)	⬆️
datafusion/core/tests/sql/group_by.rs	96.91% <100.00%> (+0.27%)	⬆️
datafusion/expr/src/logical_plan/builder.rs	90.35% <100.00%> (ø)
datafusion/core/tests/provider_filter_pushdown.rs	70.45% <0.00%> (-14.48%)	⬇️
datafusion/core/src/physical_plan/metrics/value.rs	86.93% <0.00%> (-0.51%)	⬇️
datafusion/core/src/datasource/view.rs	86.54% <0.00%> (-0.08%)	⬇️
datafusion/expr/src/function.rs	97.75% <0.00%> (ø)
datafusion/core/src/datasource/file_format/csv.rs	98.91% <0.00%> (ø)
datafusion/physical-expr/src/expressions/column.rs	100.00% <0.00%> (ø)
...fusion/optimizer/src/pre_cast_lit_in_comparison.rs	91.95% <0.00%> (ø)
... and 16 more

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

[Dandandan]

Looks great to me

[andygrove]

Thanks @isidentical and @Dandandan

[ursabot]

Benchmark runs are scheduled for baseline = 7dedcb1 and contender = 256ea91. 256ea91 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