fix(uploads): handle split-bold headings and ** ** artefacts in extract_outline (#1838)

* feat(uploads): guide agent to use grep/glob/read_file for uploaded documents

Add workflow guidance to the <uploaded_files> context block so the agent
knows to use grep and glob (added in #1784) alongside read_file when
working with uploaded documents, rather than falling back to web search.

This is the final piece of the three-PR PDF agentic search pipeline:
- PR1 (#1727): pymupdf4llm converter produces structured Markdown with headings
- PR2 (#1738): document outline injected into agent context with line numbers
- PR3 (this):  agent guided to use outline + grep + read_file workflow

* feat(uploads): add file-first priority and fallback guidance to uploaded_files context

* fix(uploads): handle split-bold headings and ** ** artefacts in extract_outline

- Add _clean_bold_title() to merge adjacent bold spans (** **) produced
  by pymupdf4llm when bold text crosses span boundaries
- Add _SPLIT_BOLD_HEADING_RE (Style 3) to recognise **<num>** **<title>**
  headings common in academic papers; excludes pure-number table headers
  and rows with more than 4 bold blocks
- When outline is empty, read first 5 non-empty lines of the .md as a
  content preview and surface a grep hint in the agent context
- Update _format_file_entry to render the preview + grep hint instead of
  silently omitting the outline section
- Add 3 new extract_outline tests and 2 new middleware tests (65 total)

* fix(uploads): address Copilot review comments on extract_outline regex

- Replace ASCII [A-Za-z] guard with negative lookahead to support non-ASCII
  titles (e.g. **1** **概述**); pure-numeric/punctuation blocks still excluded
- Replace .+ with [^*]+ and cap repetition at {0,2} (four blocks total) to
  keep _SPLIT_BOLD_HEADING_RE linear and avoid ReDoS on malformed input
- Remove now-redundant len(blocks) <= 4 code-level check (enforced by regex)
- Log debug message with exc_info when preview extraction fails

backend/tests/test_file_conversion.py

@@ -420,3 +420,40 @@ class TestExtractOutline:
         )
         outline = extract_outline(md)
         assert outline == []
+
+    def test_split_bold_heading_academic_paper(self, tmp_path):
+        """**<num>** **<title>** lines from academic papers are recognised (Style 3)."""
+        md = tmp_path / "paper.md"
+        md.write_text(
+            "## **Attention Is All You Need**\n\n**1** **Introduction**\n\nBody text.\n\n**2** **Background**\n\nMore text.\n\n**3.1** **Encoder and Decoder Stacks**\n",
+            encoding="utf-8",
+        )
+        outline = extract_outline(md)
+        titles = [e["title"] for e in outline]
+        assert "1 Introduction" in titles
+        assert "2 Background" in titles
+        assert "3.1 Encoder and Decoder Stacks" in titles
+
+    def test_split_bold_year_columns_excluded(self, tmp_path):
+        """Financial table headers like **2023** **2022** **2021** are NOT headings."""
+        md = tmp_path / "annual.md"
+        md.write_text(
+            "# Financial Summary\n\n**2023** **2022** **2021**\n\nRevenue 100 90 80\n",
+            encoding="utf-8",
+        )
+        outline = extract_outline(md)
+        titles = [e["title"] for e in outline]
+        # Only the # heading should appear, not the year-column row
+        assert titles == ["Financial Summary"]
+
+    def test_adjacent_bold_spans_merged_in_markdown_heading(self, tmp_path):
+        """** ** artefacts inside a # heading are merged into clean plain text."""
+        md = tmp_path / "sec.md"
+        md.write_text(
+            "## **UNITED STATES** **SECURITIES AND EXCHANGE COMMISSION**\n\nBody text.\n",
+            encoding="utf-8",
+        )
+        outline = extract_outline(md)
+        assert len(outline) == 1
+        # Title must be clean — no ** ** artefacts
+        assert outline[0]["title"] == "UNITED STATES SECURITIES AND EXCHANGE COMMISSION"

backend/tests/test_uploads_middleware_core_logic.py

@@ -290,6 +290,7 @@ class TestBeforeAgent:
                 "path": "/mnt/user-data/uploads/notes.txt",
                 "extension": ".txt",
                 "outline": [],
+                "outline_preview": [],
             }
         ]
 
@@ -429,3 +430,41 @@ class TestBeforeAgent:
         content = result["messages"][-1].content
         assert "Chapter 1" in content
         assert "Chapter 2" in content
+
+    def test_fallback_preview_shown_when_outline_empty(self, tmp_path):
+        """When .md exists but has no headings, first lines are shown as a preview."""
+        mw = _middleware(tmp_path)
+        uploads_dir = _uploads_dir(tmp_path)
+        (uploads_dir / "report.pdf").write_bytes(b"%PDF fake")
+        # .md with no # headings — plain prose only
+        (uploads_dir / "report.md").write_text(
+            "Annual Financial Report 2024\n\nThis document summarises key findings.\n\nRevenue grew by 12%.\n",
+            encoding="utf-8",
+        )
+
+        msg = _human("analyse", files=[{"filename": "report.pdf", "size": 9, "path": "/mnt/user-data/uploads/report.pdf"}])
+        result = mw.before_agent(self._state(msg), _runtime())
+
+        assert result is not None
+        content = result["messages"][-1].content
+        # Outline section must NOT appear
+        assert "Document outline" not in content
+        # Preview lines must appear
+        assert "Annual Financial Report 2024" in content
+        assert "No structural headings detected" in content
+        # grep hint must appear
+        assert "grep" in content
+
+    def test_fallback_grep_hint_shown_when_no_md_file(self, tmp_path):
+        """Files with no sibling .md still get the grep hint (outline is empty)."""
+        mw = _middleware(tmp_path)
+        uploads_dir = _uploads_dir(tmp_path)
+        (uploads_dir / "data.csv").write_bytes(b"a,b,c\n1,2,3\n")
+
+        msg = _human("analyse", files=[{"filename": "data.csv", "size": 12, "path": "/mnt/user-data/uploads/data.csv"}])
+        result = mw.before_agent(self._state(msg), _runtime())
+
+        assert result is not None
+        content = result["messages"][-1].content
+        assert "Document outline" not in content
+        assert "grep" in content