Refactor regexs for all test statistics and effect sizes with known acceptable formats

[fsingletonthorn]

Rewrite into individual functions for each test statistic, each returning a tibble with a column for

• Statistic
• raw reported text test stat/ES
• cleaned text stat/ES
• Test stat/ES value
• df1 (which should just contain NA for those with only 1 df)
• df2 (with should just contain the df for those with only 1 df)
• p (or NA if no p value is reported)
• context (a column with an adjustable width containing the character +- contextSize)

[fsingletonthorn]

Done