Perf: Investigate and optimize performance of Post-import Best Practices rules query execution

[PiJoCoder]

No description provided.

[PiJoCoder]

Discovered that this query can be very long if tblTopSqlPlan is populated with a large query plans. The query is processing XML in memory. The tblTopSqlPlan contains only 5 rows based on PSSDIAG collection but again if the plans are large, it is expensive to process the XML. In my test with a data set, it took slightly over 3 minutes to run this query.

set QUOTED_IDENTIFIER on; 
WITH XMLNAMESPACES ('http://schemas.microsoft.com/sqlserver/2004/07/showplan' AS sp)  
select distinct  stmt.stmt_details.value ('@Database', 'varchar(max)') 'Database' ,  stmt.stmt_details.value ('@Schema', 'varchar(max)') 'Schema' ,  
 stmt.stmt_details.value ('@Table', 'varchar(max)') 'table'   --into tblObjectsUsedByTopPlans

 from 
 (  select cast(FileContent as xml) sqlplan from tblTopSqlPlan) as p       cross apply sqlplan.nodes('//sp:Object') as stmt (stmt_details) 

[PiJoCoder]

• Also found some rules to take a long time usp_IOAnalysis due to sub-optimal logic.
• Also the update CounterData set CounterDateTime = REPLACE(CounterDateTime, char(0), '') which attempted to fix Relog.exe imports a CounterDateTime data incorrectly (likely a bug): we can work around it by updating the data manually in post-import logic #150 was not efficient. Made it more focused.