Add benchmark for parquet reader with row_filter and project settings (#7401)

* Add benchmark for parquet reader with row_filter and project settings

* fix clippy

* change bench mark to use asyn read to trigger the page cache

* fix

* fix

* Update comments, add background

* incremently addressing the comments

* Fix bool random

* fixup

* Add fn switch and project enum

* Fix clippy

* Address comment

* Add float(half set) and int(full set) change

* Address comments

* Set compression

* fix

* Update comments

* refactor filter column indexes

* Read from in memory buffer

* celanu

* Test both sync and async readers

---------

Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>

Author: zhuqi-lucas

parquet/Cargo.toml

@@ -74,7 +74,7 @@ ring = { version = "0.17", default-features = false, features = ["std"], optiona
 
 [dev-dependencies]
 base64 = { version = "0.22", default-features = false, features = ["std"] }
-criterion = { version = "0.5", default-features = false }
+criterion = { version = "0.5", default-features = false, features = ["async_futures"]  }
 snap = { version = "1.0", default-features = false }
 tempfile = { version = "3.0", default-features = false }
 brotli = { version = "7.0", default-features = false, features = ["std"] }
@@ -211,6 +211,10 @@ name = "arrow_statistics"
 required-features = ["arrow"]
 harness = false
 
+[[bench]]
+name = "arrow_reader_row_filter"
+required-features = ["arrow", "async"]
+harness = false
 
 [[bench]]
 name = "compression"