feat(F-09): rich parsing — formula + figure extraction

- codex/parsing/mathpix.py: pix2tex (local, CPU) primary + MathPix API
  optional; bbox heuristic h>15px, math-char-count>5; singleton model cache
- codex/parsing/figures.py: pymupdf embedded-image extraction → PNG;
  caption detection via proximity + "Figure/Fig./Abbildung" prefix
- codex/models.py: FormulaChunk + FigureChunk dataclasses (R-10/R-11)
- codex/ingest.py: --rich flag wires formula+figure extraction post-ingest
- codex/cli.py: search_app sub-typer (paper + formula subcommands),
  --rich flag on ingest; wiki_app from F-12 preserved intact
- codex/config.py: mathpix_app_id/key, pix2tex_fallback, figures_dir
- infra/schema.sql: formulas + figures tables with HNSW pgvector indexes
- pyproject.toml: pymupdf>=1.24, pix2tex>=0.1.4
- tests/parsing/test_mathpix.py + test_figures.py: 31 tests (mock pix2tex
  + MathPix HTTP, real pymupdf on synthetic PDF)

Gate: 158 passed, ruff clean, mypy clean (20 files)
Requirements: R-10 R-11 R-12 R-13 R-14 → done

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

codex/ingest.py

@@ -9,6 +9,7 @@ from pathlib import Path
 
 import numpy as np
 
+from codex.config import get_settings
 from codex.db import get_conn
 from codex.embed import get_embedder
 from codex.models import Citation, Paper
@@ -22,11 +23,14 @@ class IngestResult:
     paper_id: str
     chunks_upserted: int
     citations_upserted: int
+    formulas_upserted: int = 0
+    figures_upserted: int = 0
 
 
 def ingest_paper(
     paper_id: str,
     source_path: str | None = None,
+    rich: bool = False,
 ) -> IngestResult:
     """Idempotent ingest of one paper.
 
@@ -42,11 +46,15 @@ def ingest_paper(
         - ``.tex``  → :func:`codex.parsing.tex.latex_to_text` + chunk
         - ``.pdf``  → :func:`codex.parsing.nougat.pdf_to_markdown` + chunk
                       + :func:`codex.parsing.grobid.extract_references` for refs
+    rich:
+        When True and *source_path* is a PDF, also extract formulas via
+        :func:`codex.parsing.mathpix.extract_formulas` and figures via
+        :func:`codex.parsing.figures.extract_figures` (F-09).
 
     Returns
     -------
     IngestResult
-        Counts of upserted chunks and citations.
+        Counts of upserted chunks, citations, formulas, and figures.
     """
     # ---------------------------------------------------------------
     # 1. Fetch metadata (OpenAlex primary, SemanticScholar fallback)
@@ -195,8 +203,59 @@ def ingest_paper(
         citations_upserted = len(merged_citations)
         conn.commit()
 
+    # ---------------------------------------------------------------
+    # 6. F-09 Rich Parsing: formulas + figures (PDF only)
+    # ---------------------------------------------------------------
+    formulas_upserted = 0
+    figures_upserted = 0
+
+    if rich and source_path is not None and Path(source_path).suffix.lower() == ".pdf":
+        from codex.parsing.figures import extract_figures
+        from codex.parsing.mathpix import extract_formulas
+
+        settings = get_settings()
+
+        formulas = extract_formulas(source_path)
+        figures = extract_figures(source_path, output_dir=settings.figures_dir)
+
+        if formulas or figures:
+            with get_conn() as conn2:
+                if formulas:
+                    with conn2.cursor() as cur:
+                        cur.executemany(
+                            """
+                            INSERT INTO formulas (paper_id, page, raw_latex, context, eq_label)
+                            VALUES (%s, %s, %s, %s, %s)
+                            ON CONFLICT DO NOTHING
+                            """,
+                            [
+                                (f.paper_id, f.page, f.raw_latex, f.context, f.eq_label)
+                                for f in formulas
+                            ],
+                        )
+                    formulas_upserted = len(formulas)
+
+                if figures:
+                    with conn2.cursor() as cur:
+                        cur.executemany(
+                            """
+                            INSERT INTO figures (paper_id, page, caption, image_path)
+                            VALUES (%s, %s, %s, %s)
+                            ON CONFLICT DO NOTHING
+                            """,
+                            [
+                                (fig.paper_id, fig.page, fig.caption, fig.image_path)
+                                for fig in figures
+                            ],
+                        )
+                    figures_upserted = len(figures)
+
+                conn2.commit()
+
     return IngestResult(
         paper_id=paper.id,
         chunks_upserted=chunks_upserted,
         citations_upserted=citations_upserted,
+        formulas_upserted=formulas_upserted,
+        figures_upserted=figures_upserted,
     )