Fix some naming related to AggregatePullUpLookupRule. (#15677)

It was called "split" rather than "pull up" in some places. This patch
standardizes on "pull up".

sql/src/main/java/org/apache/druid/sql/calcite/expression/builtin/QueryLookupOperatorConversion.java

@@ -90,14 +90,14 @@ public DruidExpression toDruidExpression(
                     false,
                     replaceMissingValueWith,
                     null,
-                    // If SplitLookupRule is disabled, then enable optimization at the extractionFn level, since a
+                    // If LOOKUP pull-up is disabled, then enable optimization at the extractionFn level, since a
                     // similar optimization may be done by the native query toolchests. We'd like to ensure that if
                     // people upgrade to a version where this rule was added, and then disable the rule due to some
                     // problem with it, they still get any optimization that the native layer was able to do.
                     //
                     // Note that we don't check plannerContext.isReverseLookup(), because the native layer doesn't
                     // optimize filters on RegisteredLookupExtractionFn anyway.
-                    !plannerContext.isSplitLookup()
+                    !plannerContext.isPullUpLookup()
                 )
             );
           } else {

sql/src/main/java/org/apache/druid/sql/calcite/planner/CalciteRulesManager.java

@@ -333,7 +333,7 @@ private Program buildReductionProgram(final PlannerContext plannerContext, final
       builder.addRuleInstance(new FilterDecomposeConcatRule());
 
       // Include rule to split injective LOOKUP across a GROUP BY.
-      if (plannerContext.isSplitLookup()) {
+      if (plannerContext.isPullUpLookup()) {
         builder.addRuleInstance(new AggregatePullUpLookupRule(plannerContext));
       }
 

sql/src/main/java/org/apache/druid/sql/calcite/planner/PlannerContext.java

@@ -96,7 +96,7 @@ public class PlannerContext
   public static final boolean DEFAULT_SQL_USE_BOUNDS_AND_SELECTORS = NullHandling.replaceWithDefault();
 
   /**
-   * Context key for {@link PlannerContext#isSplitLookup()}.
+   * Context key for {@link PlannerContext#isPullUpLookup()}.
    */
   public static final String CTX_SQL_PULL_UP_LOOKUP = "sqlPullUpLookup";
   public static final boolean DEFAULT_SQL_PULL_UP_LOOKUP = true;
@@ -120,7 +120,7 @@ public class PlannerContext
   private final String sqlQueryId;
   private final boolean stringifyArrays;
   private final boolean useBoundsAndSelectors;
-  private final boolean splitLookup;
+  private final boolean pullUpLookup;
   private final boolean reverseLookup;
   private final CopyOnWriteArrayList<String> nativeQueryIds = new CopyOnWriteArrayList<>();
   private final PlannerHook hook;
@@ -148,7 +148,7 @@ private PlannerContext(
       final DateTime localNow,
       final boolean stringifyArrays,
       final boolean useBoundsAndSelectors,
-      final boolean splitLookup,
+      final boolean pullUpLookup,
       final boolean reverseLookup,
       final SqlEngine engine,
       final Map<String, Object> queryContext,
@@ -164,7 +164,7 @@ private PlannerContext(
     this.localNow = Preconditions.checkNotNull(localNow, "localNow");
     this.stringifyArrays = stringifyArrays;
     this.useBoundsAndSelectors = useBoundsAndSelectors;
-    this.splitLookup = splitLookup;
+    this.pullUpLookup = pullUpLookup;
     this.reverseLookup = reverseLookup;
     this.hook = hook == null ? NoOpPlannerHook.INSTANCE : hook;
 
@@ -188,14 +188,14 @@ public static PlannerContext create(
     final DateTimeZone timeZone;
     final boolean stringifyArrays;
     final boolean useBoundsAndSelectors;
-    final boolean splitLookup;
+    final boolean pullUpLookup;
     final boolean reverseLookup;
 
     final Object stringifyParam = queryContext.get(QueryContexts.CTX_SQL_STRINGIFY_ARRAYS);
     final Object tsParam = queryContext.get(CTX_SQL_CURRENT_TIMESTAMP);
     final Object tzParam = queryContext.get(CTX_SQL_TIME_ZONE);
     final Object useBoundsAndSelectorsParam = queryContext.get(CTX_SQL_USE_BOUNDS_AND_SELECTORS);
-    final Object splitLookupParam = queryContext.get(CTX_SQL_PULL_UP_LOOKUP);
+    final Object pullUpLookupParam = queryContext.get(CTX_SQL_PULL_UP_LOOKUP);
     final Object reverseLookupParam = queryContext.get(CTX_SQL_REVERSE_LOOKUP);
 
     if (tsParam != null) {
@@ -222,10 +222,10 @@ public static PlannerContext create(
       useBoundsAndSelectors = DEFAULT_SQL_USE_BOUNDS_AND_SELECTORS;
     }
 
-    if (splitLookupParam != null) {
-      splitLookup = Numbers.parseBoolean(splitLookupParam);
+    if (pullUpLookupParam != null) {
+      pullUpLookup = Numbers.parseBoolean(pullUpLookupParam);
     } else {
-      splitLookup = DEFAULT_SQL_PULL_UP_LOOKUP;
+      pullUpLookup = DEFAULT_SQL_PULL_UP_LOOKUP;
     }
 
     if (reverseLookupParam != null) {
@@ -241,7 +241,7 @@ public static PlannerContext create(
         utcNow.withZone(timeZone),
         stringifyArrays,
         useBoundsAndSelectors,
-        splitLookup,
+        pullUpLookup,
         reverseLookup,
         engine,
         queryContext,
@@ -378,13 +378,12 @@ public boolean isUseBoundsAndSelectors()
   }
 
   /**
-   * Whether we should use {@link AggregatePullUpLookupRule} to split LOOKUP functions on injective lookups when they
-   * are dimensions in aggregations, and whether we should set the "optimize" flag on
-   * {@link RegisteredLookupExtractionFn}.
+   * Whether we should use {@link AggregatePullUpLookupRule} to pull LOOKUP functions on injective lookups up above
+   * a GROUP BY.
    */
-  public boolean isSplitLookup()
+  public boolean isPullUpLookup()
   {
-    return splitLookup;
+    return pullUpLookup;
   }
 
   /**