Add support for "per table" settings for stats update

[yochananrachamim]

for scenario where we have large tables and we do not want to go through all the data for the stats update it would help to have some control for per table settings that will set the percentage of the scan while updating stats.

[sopelt]

This would also be nice for index operations. In niche cases (large tables that are just appended to and rarely read) excluding by table would be great.

[akobuspl]

I'd vote for possibility to exclude either a specific table or specific statistics.
It happens that some column statistics which are not having big value are causing statistics update takes "forever"

[sopelt]

Hi @yochananrachamim , this seems to be implemented in the more recent versions. Is it possible, that there are still issues with "uncommon" collations? e.g. if the collation is case sensitive it looks like the checks in

AzureSQL/AzureSQLMaintenance.sql

Line 414 in c4598b0

on ovr.ApplyOnObjectType='Schema' and ovr.IsSample=0 and ovr.Operation='Index' and ovr.OverrideName='Exclude'

vs.

AzureSQL/AzureSQLMaintenance.sql

Line 614 in c4598b0

on ovr.ApplyOnObjectType='schema' and ovr.IsSample=0 and ovr.Operation = 'Statistics' and ovr.OverrideName='exclude'

might match differently. This could apply to ApplyOnObjectType, Operation, OverrideName, TableSchema, TableName, StatisticsName if I am not missing something.

Thanks for the continued work on the script!