fix: document upsert project_id, truncation budget, and Ollama model matching

- regenerator: Include project_id in the ON CONFLICT UPDATE clause for
  document upserts. Previously, if a document moved between projects
  (e.g., during re-ingestion), the project_id would remain stale.

- truncation: Compute the omission marker ("N notes omitted") before
  checking whether first+last notes fit in the budget. The old order
  computed the marker after the budget check, meaning the marker's byte
  cost was unaccounted for and could cause over-budget output.

- ollama: Tighten model name matching to require either an exact match
  or a colon-delimited tag prefix (model == name or name starts with
  "model:"). The prior starts_with check would false-positive on
  "nomic-embed-text-v2" when looking for "nomic-embed-text". Tests
  updated to cover exact match, tagged, wrong model, and prefix
  false-positive cases.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/documents/truncation.rs

@@ -110,7 +110,9 @@ pub fn truncate_discussion(notes: &[NoteContent], max_bytes: usize) -> Truncatio
     }
 
     let first_note = &formatted[0];
-    if first_note.len() + last_note.len() > max_bytes {
+    let omitted = formatted.len() - 2;
+    let marker = format!("\n\n[... {} notes omitted for length ...]\n\n", omitted);
+    if first_note.len() + marker.len() + last_note.len() > max_bytes {
         let truncated = truncate_utf8(first_note, max_bytes.saturating_sub(11));
         let content = format!("{}[truncated]", truncated);
         return TruncationResult {
@@ -120,8 +122,6 @@ pub fn truncate_discussion(notes: &[NoteContent], max_bytes: usize) -> Truncatio
         };
     }
 
-    let omitted = formatted.len() - 2;
-    let marker = format!("\n\n[... {} notes omitted for length ...]\n\n", omitted);
     let content = format!("{}{}{}", formatted[0], marker, last_note);
     TruncationResult {
         content,