upgrade to RoaringBitmap 0.9.25

[richardstartin]

Picks up some performance improvements in RangeBitmap:

Benchmark                    (_dataType)  (_decile)  (_numDocs)                          (_scenario)  (_seed)  Mode  Cnt      Score     Error  Units
BenchmarkRangeIndex.queryV2         LONG          5    10000000                          NORMAL(0,1)       42  avgt    5   1506.798 ±  35.190  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                NORMAL(10000000,1000)       42  avgt    5   5558.664 ± 125.293  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                          EXP(0.0001)       42  avgt    5   8785.747 ± 649.172  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                             EXP(0.5)       42  avgt    5   3226.282 ±  11.844  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000              UNIFORM(0,100000000000)       42  avgt    5  13849.725 ± 721.279  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000  UNIFORM(100000000000, 100000000100)       42  avgt    5   2097.835 ±  54.656  us/op

Benchmark                    (_dataType)  (_decile)  (_numDocs)                          (_scenario)  (_seed)  Mode  Cnt      Score      Error  Units
BenchmarkRangeIndex.queryV2         LONG          5    10000000                          NORMAL(0,1)       42  avgt    5   1071.426 ±   64.424  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                NORMAL(10000000,1000)       42  avgt    5   4531.646 ±   66.475  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                          EXP(0.0001)       42  avgt    5   6177.665 ±  110.414  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000                             EXP(0.5)       42  avgt    5   3113.483 ±   16.219  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000              UNIFORM(0,100000000000)       42  avgt    5  10834.696 ± 3089.046  us/op
BenchmarkRangeIndex.queryV2         LONG          5    10000000  UNIFORM(100000000000, 100000000100)       42  avgt    5   1868.769 ±   39.611  us/op

Also memory mapping a bitmap is now ~25% faster.

[codecov-commenter]

Codecov Report

Merging #8206 (0f22b1d) into master (df39bda) will decrease coverage by 3.62%.
The diff coverage is 75.16%.

❗ Current head 0f22b1d differs from pull request most recent head 36bb8b0. Consider uploading reports for the commit 36bb8b0 to get more accurate results

@@             Coverage Diff              @@
##             master    #8206      +/-   ##
============================================
- Coverage     71.01%   67.38%   -3.63%     
+ Complexity     4314     4232      -82     
============================================
  Files          1624     1226     -398     
  Lines         84873    61895   -22978     
  Branches      12791     9631    -3160     
============================================
- Hits          60273    41709   -18564     
+ Misses        20453    17222    -3231     
+ Partials       4147     2964    -1183     

Flag	Coverage Δ
integration1	?
integration2	?
unittests1	67.38% <75.16%> (-0.07%)	⬇️
unittests2	?

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

Impacted Files	Coverage Δ
...he/pinot/core/api/ServiceAutoDiscoveryFeature.java	0.00% <0.00%> (ø)
...va/org/apache/pinot/spi/utils/CommonConstants.java	23.63% <ø> (ø)
...sform/function/BaseBinaryGeoTransformFunction.java	54.83% <54.83%> (ø)
...spatial/transform/function/StDistanceFunction.java	85.71% <57.14%> (+1.23%)	⬆️
...spatial/transform/function/StContainsFunction.java	57.14% <60.00%> (-18.72%)	⬇️
...eospatial/transform/function/StWithinFunction.java	57.14% <60.00%> (-18.72%)	⬇️
...or/transform/function/LookupTransformFunction.java	79.05% <73.52%> (-9.06%)	⬇️
...inot/core/query/reduce/PostAggregationHandler.java	91.76% <83.33%> (+0.09%)	⬆️
...upby/NoDictionaryMultiColumnGroupKeyGenerator.java	63.55% <94.28%> (+0.62%)	⬆️
...eospatial/transform/function/StEqualsFunction.java	80.00% <100.00%> (-2.15%)	⬇️
... and 629 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 df39bda...36bb8b0. Read the comment docs.