[fix](iceberg) fix iceberg count(*) short circuit read bug

[wuwenchi]

Proposed changes

When using the nereids optimizer, the count operation needs to be pushed down before fetching split.
So the current order is:

1. create a ScanNode
2. set the filter
3. set the push down operation of the ScanNode, such as count
4. generate split for this ScanNode

Further comments

If this is a relatively large or complex change, kick off the discussion at [email protected] by explaining why you chose the solution you did and what alternatives you considered, etc...

[wuwenchi]

run buildall

[morningman]

LGTM
@morrySnow PTAL

[github-actions]

PR approved by at least one committer and no changes requested.

[github-actions]

PR approved by anyone and no changes requested.

[morningman]

run buildall

[hello-stephen]

(From new machine)TeamCity pipeline, clickbench performance test result:
the sum of best hot time: 46.92 seconds
stream load tsv: 541 seconds loaded 74807831229 Bytes, about 131 MB/s
stream load json: 21 seconds loaded 2358488459 Bytes, about 107 MB/s
stream load orc: 66 seconds loaded 1101869774 Bytes, about 15 MB/s
stream load parquet: 31 seconds loaded 861443392 Bytes, about 26 MB/s
insert into select: 29.4 seconds inserted 10000000 Rows, about 340K ops/s
storage size: 17162038574 Bytes

[wuwenchi]

run buildall

[wuwenchi]

run buildall

[hello-stephen]

(From new machine)TeamCity pipeline, clickbench performance test result:
the sum of best hot time: 46.28 seconds
stream load tsv: 539 seconds loaded 74807831229 Bytes, about 132 MB/s
stream load json: 20 seconds loaded 2358488459 Bytes, about 112 MB/s
stream load orc: 65 seconds loaded 1101869774 Bytes, about 16 MB/s
stream load parquet: 31 seconds loaded 861443392 Bytes, about 26 MB/s
insert into select: 29.4 seconds inserted 10000000 Rows, about 340K ops/s
storage size: 17162210277 Bytes

[wuwenchi]

run buildall

[hello-stephen]

(From new machine)TeamCity pipeline, clickbench performance test result:
the sum of best hot time: 46.66 seconds
stream load tsv: 542 seconds loaded 74807831229 Bytes, about 131 MB/s
stream load json: 20 seconds loaded 2358488459 Bytes, about 112 MB/s
stream load orc: 65 seconds loaded 1101869774 Bytes, about 16 MB/s
stream load parquet: 32 seconds loaded 861443392 Bytes, about 25 MB/s
insert into select: 29.3 seconds inserted 10000000 Rows, about 341K ops/s
storage size: 17162285250 Bytes

[xiaokang]

LGTM