⚡ refactor tableRuns to better manage memory

[bunop]

Is your feature request related to a problem? Please describe.
tableRuns consume a lot of memory especially when applied to WGS data

Describe the solution you'd like
Rewrite function to manage memory better, for example calculating data by chromosomes without store values in memory

Describe alternatives you've considered
Since SNPs in runs below the supplied threshold are discarded, we could filter directly those SNPs in snpInsideRunsCpp

Additional context

• add test code for tableRuns
• remove SnpInRuns as input parameter (snpInsideRuns is now internal)
• filter SNPs after snpInsideRunCpp or in snpInsideRunsCpp
• apply all the steps by chromosomes
• implement tableRuns in cpp
• remove unused R code
• readPOPcpp once and apply the same pop object to snpInsideRunCpp