Fix ANALYZE when Hive partition has non-canonical value

[denodo-research-labs]

Description

Extracted from trinodb/trino#15995

In Hive it may well happen that a partition value is written by the writer process as a string,
e.g. : month=02, even though the column is registered in Hive as an integer.

When updating the table or when doing ANALYZE, the output in Presto of the statistics computation though for the partition
from the example above will be though 2, ending in the the following error: All computed statistics must be used.

Motivation and Context

While performing ANALYZE on the following partitioned dataset:
store_sales/d_year=2025/d_month=01/d_day=10/d_hour=00

the following exception occurs:

com.google.common.base.VerifyException: All computed statistics must be used
at com.google.common.base.Verify.verify(Verify.java:126)
at com.facebook.presto.hive.HiveMetadata.finishStatisticsCollection(HiveMetadata.java:1552)
at com.facebook.presto.spi.connector.classloader.ClassLoaderSafeConnectorMetadata.finishStatisticsCollection(ClassLoaderSafeConnectorMetadata.java:210)
at com.facebook.presto.metadata.MetadataManager.finishStatisticsCollection(MetadataManager.java:761)

This PR addresses the above mentioned issue by parsing the partition values to Presto values in order to avoid ignoring computed statistics.

Test Plan

Added test method testAnalyzePartitionedTableWithNonCanonicalValues

Contributor checklist

• Please make sure your submission complies with our contributing guide, in particular code style and commit standards.
• PR description addresses the issue accurately and concisely. If the change is non-trivial, a GitHub Issue is referenced.
• If release notes are required, they follow the release notes guidelines.
• Adequate tests were added if applicable.
• CI passed.

Release Notes

Please follow release notes guidelines and fill in the release notes below.

== RELEASE NOTES ==

Hive Connector Changes
* Fix incorrectly ignoring computed table statistics in `ANALYZE`