[FEATURE]PPL Support approximation count for improved performance

[YANG-DB]

Is your feature request related to a problem?
In large volume datasets and/or high cardinality datasets the counting is an expensive computation.
Spark has a list of build-in approximation function such as:

• approx_count_distinct which returns the estimated number of distinct values in expr within the group.

  •   SELECT approx_count_distinct(column_name) FROM table_name;

• approx_percentile which returns the approximate percentile of the numeric column col which is the smallest value in the ordered col values

  •   SELECT 
         col_name,
             approx_percentile(col_name, 0.5) AS median_value,
             approx_percentile(col_name, array(0.25, 0.5, 0.75)) AS quartiles
         FROM 
             table_name
         GROUP BY 
             col_name

What solution would you like?
We would like to add these approximation capabilities to every PPL function that can offer counting:

• Top & Rare commands

... | top_approx 5 values by country 
... | rare_approx age by gender

• Stats

... | stats count_distinct_approx(c) by b | head 5` 

Do you have any additional context?

• https://spark.apache.org/docs/3.5.2/sql-ref-functions-builtin.html