Restore the unconditional block on system command execution

ClaudeR 0.12.6. A local experiment left validate_code_security gating the
system/system2/shell block behind getOption("clauder.allow_system"), and
that gate reached a public release. Setting an R option is ordinary code
that the validator does not inspect, so an agent could switch the option
on through execute_r and then run shell commands: the restriction became
advisory rather than enforced. The block is unconditional again and there
is no option that lifts it.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

Author: IMNMV

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: ClaudeR
 Title: R Integration for Claude AI
-Version: 0.12.5
+Version: 0.12.6
 Authors@R: person("Nykko", "Vitali", email = "nykvt@icloud.com", role = c("aut", "cre"))
 Description: Connects RStudio with Claude AI to enable interactive coding sessions.
 License: MIT + file LICENSE

R/ui.R

@@ -1717,11 +1717,11 @@ query_agent_history <- function(agent_filter = "all", requesting_agent = NULL,
 #' @return A list with blocked (logical) and reason (character) fields
 
 validate_code_security <- function(code) {
-  # System command calls: allowed only when getOption('clauder.allow_system', FALSE) is TRUE (default off = public shipped behavior)
-  if (!isTRUE(getOption("clauder.allow_system", FALSE)) && (grepl("\\bsystem\\s*\\(", code) ||
+  # System command calls to block completely
+  if (grepl("\\bsystem\\s*\\(", code) ||
       grepl("\\bsystem2\\s*\\(", code) ||
       grepl("\\bshell\\s*\\(", code) ||
-      grepl("\\bshell\\.exec\\s*\\(", code))) {
+      grepl("\\bshell\\.exec\\s*\\(", code)) {
     return(list(
       blocked = TRUE,
       reason = "Security restriction: System command execution is not allowed"