ESQL: Explain confusing query (#135848) (#136298)

This links to an explanation about a very confusing query that you can
write in ESQL. It've decided to link to github here to because this
query is so deeply at the edge that you need to know that you are
dealing with hackers now.

Closes #134792

Author: nik9000

docs/reference/query-languages/esql/_snippets/commands/layout/stats-by.md

@@ -75,7 +75,7 @@ and then grouping - that is not going to be faster.
 ::::
 
 
-**Examples**
+### Examples
 
 Calculating a statistic and grouping by the values of another column:
 
@@ -104,15 +104,17 @@ optional as well:
 :::{include} ../examples/stats.csv-spec/aggFilteringNoGroup.md
 :::
 
-$$$esql-stats-mv-group$$$
-If the grouping key is multivalued then the input row is in all groups:
+It’s also possible to group by multiple values:
 
-:::{include} ../examples/stats.csv-spec/mv-group.md
+:::{include} ../examples/stats.csv-spec/statsGroupByMultipleValues.md
 :::
 
-It’s also possible to group by multiple values:
+$$$esql-stats-mv-group$$$
+#### Multivalued inputs
 
-:::{include} ../examples/stats.csv-spec/statsGroupByMultipleValues.md
+If the grouping key is multivalued then the input row is in all groups:
+
+:::{include} ../examples/stats.csv-spec/mv-group.md
 :::
 
 If all the grouping keys are multivalued then the input row is in all groups:
@@ -132,6 +134,11 @@ key. If you want to send the group key to the function then `MV_EXPAND` first:
 :::{include} ../examples/stats.csv-spec/mv-group-values-expand.md
 :::
 
+Refer to [elasticsearch/issues/134792](https://github.com/elastic/elasticsearch/issues/134792#issuecomment-3361168090)
+for an even more in depth explanation.
+
+#### Multivalue functions
+
 Both the aggregating functions and the grouping expressions accept other
 functions. This is useful for using `STATS` on multivalue columns.
 For example, to calculate the average salary change, you can use `MV_AVG` to
@@ -147,6 +154,8 @@ letter of their last name:
 :::{include} ../examples/stats.csv-spec/docsStatsByExpression.md
 :::
 
+#### Naming
+
 Specifying the output column name is optional. If not specified, the new column
 name is equal to the expression. The following query returns a column named
 `AVG(salary)`: