Optimized sql for Cohort Era. Thank you @chrisknoll who wrote the SQL
Explicitly cast cohort_definition_id, subject_id as BIGINT and cohort_start_date, cohort_end_date as DATE. This reduces the chances of cohort table having INT or date-time.
Added new function reindexRules(). Tests added. Tests pass. Improved test coverage.
Added new function called reindexCohortsByDays()
Improved test coverage.
Made CRAN compatible.
Made CRAN compatible.
Removed experimental functions.
This is a unreleased package.