Ensure shell command skills trigger approval (#12697)

Summary
- detect skill-invoking shell commands based on the original command
string, request approvals when needed, and cache positive decisions per
session
- keep implicit skill invocation emitted after approval and keep skill
approval decline messaging centralized to the shell handler
- expand and adjust skill approval tests to cover shell-based skill
scripts while matching the new detection expectations

Testing
- Not run (not requested)

codex-rs/linux-sandbox/tests/suite/landlock.rs

@@ -73,6 +73,7 @@ async fn run_cmd_result_with_writable_roots(
     let sandbox_cwd = cwd.clone();
     let params = ExecParams {
         command: cmd.iter().copied().map(str::to_owned).collect(),
+        original_command: cmd.iter().copied().map(str::to_owned).collect(),
         cwd,
         expiration: timeout_ms.into(),
         env: create_env_from_core_vars(),
@@ -315,6 +316,7 @@ async fn assert_network_blocked(cmd: &[&str]) {
     let sandbox_cwd = cwd.clone();
     let params = ExecParams {
         command: cmd.iter().copied().map(str::to_owned).collect(),
+        original_command: cmd.iter().copied().map(str::to_owned).collect(),
         cwd,
         // Give the tool a generous 2-second timeout so even slow DNS timeouts
         // do not stall the suite.