fix nested column thread safety issue for 24.0.1

[clintropolis]

smaller footprint version of fix #13265 for 24.0.1, this just fixes the TypeStrategy to use positional reads

[imply-cheddar]

This change is just general refactoring and not actually functionally meaningful, right? Just validating my understanding

[clintropolis]

yep, made writing the test easier and the direct field access was kind of ugly

[xvrl]

@clintropolis it looks like integration tests are still failing in the 24.0.1 branch. We might want to check if there is something non-deterministic or not.

[clintropolis]

it looks like integration tests are still failing in the 24.0.1 branch. We might want to check if there is something non-deterministic or not.

hmm, these actually look unrelated to my fix, but maybe worth looking into

[ERROR] org.apache.druid.sql.avatica.DruidAvaticaHandlerTest.testConcurrentQueries  Time elapsed: 12.981 s  <<< ERROR!
java.util.concurrent.ExecutionException: java.lang.RuntimeException: org.apache.calcite.avatica.AvaticaSqlException: Error -1 (00000) : Error while executing SQL "SELECT COUNT(*) + 1636 AS ci FROM foo": Remote driver error: QueryInterruptedException: Error while applying rule DruidQueryRule(AGGREGATE_PROJECT), args [rel#631993:LogicalProject.NONE.[](input=RelSubset#631992,ci=+($0, 1636)), rel#632012:DruidQueryRel.NONE.[](query={"queryType":"timeseries","dataSource":{"type":"table","name":"foo"},"intervals":{"type":"intervals","intervals":["-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z"]},"granularity":{"type":"all"},"aggregations":[{"type":"count","name":"a0"}],"context":{"nativeQueryIds":"[3ef1e7e1-f2f6-4152-8384-229eaea39482]","sqlQueryId":"f2727102-ff6c-404c-8f9a-2ed84a406e46","sqlStringifyArrays":false}},signature={a0:LONG})] -> RuntimeException: Error while applying rule DruidQueryRule(AGGREGATE_PROJECT), args [rel#631993:LogicalProject.NONE.[](input=RelSubset#631992,ci=+($0, 1636)), rel#632012:DruidQueryRel.NONE.[](query={"queryType":"timeseries","dataSource":{"type":"table","name":"foo"},"intervals":{"type":"intervals","intervals":["-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z"]},"granularity":{"type":"all"},"aggregations":[{"type":"count","name":"a0"}],"context":{"nativeQueryIds":"[3ef1e7e1-f2f6-4152-8384-229eaea39482]","sqlQueryId":"f2727102-ff6c-404c-8f9a-2ed84a406e46","sqlStringifyArrays":false}},signature={a0:LONG})] -> NullPointerException: Cannot invoke "java.util.Map.size()" because "m" is null

I think maybe #13049 is the fix

[clintropolis]

created #13281 to backport that fix