GCS: bump up gcs-analytics-core version from 1.2.1 -> 1.2.3

[ajayky-os]

Changelog of gcs-analytics-core:

• Fix issue with vectored IO where read requests were not bounded, resulting in poor performance with larger data files.
• Disable small object prefetch optimization by default, does not contribute to performance gain for larger data set, regresses few querie's scan time.

Consecutive Benchmark Results with updated version for 1 TB schema size:

benchmark	scan_time (analytics core disabled)	scan_time (analytics core enabled)	% Scan time improvement
tpcds_sf1000	29,946,605	19,370,576	35.32
tpcds_sf1000	29,626,952	18,743,899	36.73
tpcds_sf1000	30,714,037	19,084,079	37.87
tpcds_sf1000	29,281,478	18,700,487	36.14

benchmark	scan_time (analytics core disabled)	scan_time (analytics core enabled)	% Scan time improvement
tpch_sf1000	16,753,626	13,833,931	17.43
tpch_sf1000	17,615,900	14,144,223	19.71
tpch_sf1000	17,780,989	13,958,524	21.50
tpch_sf1000	17,025,561	13,746,540	19.26

*scan_time is total of custom scan time metric added in BatchScan nodes for all queries. This was achieved by adding the metric in BatchDataReader in Iceberg spark runtime and captured the metrics by registering SparkListener in spark job running queries.

[huaxingao]

Thanks @ajayky-os for the PR!