[FEA] Support map_filter

[viadea]

I wish we can support map_filter function.

For example:

from pyspark.sql.types import *
from pyspark.sql.functions import *
jsonString="""{"Zipcode":704,"ZipCodeType":"STANDARD","City":"PARC PARQUE","State":"PR"}"""
df=spark.createDataFrame([(1, jsonString)],["id","value"])
df.show(truncate=False)
df2=df.withColumn("value",from_json(df.value,MapType(StringType(),StringType())))

spark.conf.set("spark.rapids.sql.regexp.enabled",True)
df3=df2.withColumn('mapfiltercol', expr(f""" map_filter(value, (k,v)  -> k rlike "Zip%"    )       """) )
df3.collect()

Not-supported-messages:

    ! <MapFilter> map_filter(from_json(MapType(StringType,StringType,true), value#12, Some(UTC)), lambdafunction(RLIKE(lambda k#70, Zip%), lambda k#70, lambda v#71, false)) cannot run on GPU because GPU does not currently support the operator class org.apache.spark.sql.catalyst.expressions.MapFilter

[sameerz]

Depends on rapidsai/cudf#10650