Add histogram aggregation function

[jasperjiaguo]

This effort is part of the issue #8493
The histogram aggregation function for single value numerical columns is implemented.
usage example:
Histogram(columnName, ARRAY[0,1,10,100]) to specify bins [0,1), [1,10), [10,1000]
HISTOGRAM(intColumn,ARRAY["-Infinity",0,1,10,100,1000, "+Infinity"]) to specify bins (-inf, 0), [0,1), [1,10), [10,1000), [1000, +inf)
Histogram(columnName, 0, 1000, 10) to specify 10 equal-length bins [0,100), [100,200), ..., [900,1000]
This is a proof of concept histogram function, the following are some TODOs that we want to address in a follow up pr:

1. Add support for multi-value column
2. Add support for automatically collect min/max values when lower/upper bounds are not specified
3. Count for outlier values as an additional bin

[codecov-commenter]

Codecov Report

Merging #8724 (fced48a) into master (d2396dd) will decrease coverage by 0.01%.
The diff coverage is 64.02%.

@@             Coverage Diff              @@
##             master    #8724      +/-   ##
============================================
- Coverage     69.68%   69.67%   -0.02%     
- Complexity     4577     4618      +41     
============================================
  Files          1729     1736       +7     
  Lines         90183    91180     +997     
  Branches      13415    13632     +217     
============================================
+ Hits          62848    63529     +681     
- Misses        22971    23226     +255     
- Partials       4364     4425      +61     

Flag	Coverage Δ
integration1	26.98% <0.00%> (-0.12%)	⬇️
integration2	25.32% <0.00%> (-0.10%)	⬇️
unittests1	66.15% <64.02%> (-0.01%)	⬇️
unittests2	14.18% <0.00%> (-0.09%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...java/org/apache/pinot/common/utils/DataSchema.java	69.00% <50.00%> (+0.25%)	⬆️
...e/query/aggregation/utils/DoubleVectorOpUtils.java	52.94% <52.94%> (ø)
...egation/function/HistogramAggregationFunction.java	64.45% <64.45%> (ø)
...gregation/function/AggregationFunctionFactory.java	81.20% <100.00%> (+0.28%)	⬆️
...che/pinot/segment/spi/AggregationFunctionType.java	89.53% <100.00%> (+0.12%)	⬆️
...g/apache/pinot/server/api/resources/ErrorInfo.java	0.00% <0.00%> (-100.00%)	⬇️
...t/server/api/resources/DefaultExceptionMapper.java	0.00% <0.00%> (-75.00%)	⬇️
...che/pinot/controller/util/TableMetadataReader.java	63.41% <0.00%> (-36.59%)	⬇️
...icate/FSTBasedRegexpPredicateEvaluatorFactory.java	56.25% <0.00%> (-33.75%)	⬇️
.../index/loader/invertedindex/RangeIndexHandler.java	49.55% <0.00%> (-17.70%)	⬇️
... and 116 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d2396dd...fced48a. Read the comment docs.

[Jackie-Jiang]

Suggest changing the function name to histogram (remove the SV). We should be able to use the same function to handle both SV and MV in the future.

[Jackie-Jiang]

Thanks for adding the feature. Please also update the pinot doc to include this new function.

[Jackie-Jiang]

Can be simplified to:

return vectorAdd(a, b.elements());

[Jackie-Jiang]

Call a.elements() first and cache the value, then operate on 2 double[] for better performance. Not sure if jvm will inline all the method calls. Same for incrementElement()

[Jackie-Jiang]

Remove