M6: a document script's error is told apart by where it came from, not its wording (#3940)

Review follow-up on #3985: JsOutput deduplicated document-script errors by
their text (document script N: ...), so a field script that threw the
same words was dropped by append, and a document script's error was
dropped by absorb when a field script had already reported that text.
JsOutput now records which errors came from document scripts
(doc_errors) and deduplicates only those; pdfcraft_js::is_doc_error,
which guessed from the wording, is removed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0175FWj4xFhMhSH3UpMU8DSC

crates/js/src/lib.rs

@@ -1274,21 +1274,9 @@ impl HostHooks for CheckedCodeFromStrings {
     }
 }
 
-/// The prefix of a document script's error ([`doc_error`]).
-const DOC_ERROR_PREFIX: &str = "document script ";
-
 /// How [`Outcome::doc_errors`] names a failing document script: by its 1-based place.
 fn doc_error(index: usize, error: &dyn std::fmt::Display) -> String {
-    format!("{DOC_ERROR_PREFIX}{}: {error}", index.saturating_add(1))
-}
-
-/// Whether `error` is one of [`Outcome::doc_errors`]: a document script's failure, which repeats
-/// each time the document scripts run again, unlike a field script's own error.
-pub fn is_doc_error(error: &str) -> bool {
-    error
-        .strip_prefix(DOC_ERROR_PREFIX)
-        .and_then(|rest| rest.split_once(": "))
-        .is_some_and(|(n, _)| !n.is_empty() && n.bytes().all(|b| b.is_ascii_digit()))
+    format!("document script {}: {error}", index.saturating_add(1))
 }
 
 /// `doc_scripts` are the document scripts to run with their places among all of them;