Fix apostrophes and commas in exclusion reasons

Problem

Users encountered two issues when entering exclusion reasons:

1. Syntax Error 140 when exclusion reasons contained apostrophes (e.g., "doesn't", "isn't", "won't")
2. Text truncation when exclusion reasons contained commas - only text before the first comma was displayed

Root Cause

1. Apostrophes/quotes: DiagrammeR's DOT syntax parser failed when unescaped special characters were inserted directly into the graph definition
2. Commas: The parsing logic used naive regex (",.*$" and ".*,") that split on the first/last comma, truncating exclusion reasons that legitimately contained commas

Solution

1. Added text escaping for DOT syntax

• Created PRISMA_escape_text_() utility function in utils.R
• Escapes backslashes, quotes, and apostrophes before inserting into DiagrammeR graph
• Applied to both dbr_excluded[, 1] and other_excluded[, 1] text

2. Fixed exclusion reason parsing logic

• Changed from naive comma splitting to pattern-based parsing
• Before: gsub(",.*$", "", text) - removed everything after first comma
• After: gsub(",\\s*\\d+\\s*$", "", text) - removes only final comma + number
• Before: gsub(".*,", "", text) - removed everything before last comma
• After: gsub("^.*,\\s*(\\d+)\\s*$", "\\1", text) - extracts only final number