repo hygiene, tooling config, v0.2.0

- Untrack src/ocr_pipeline.egg-info from git, delete pyproject.toml~,
  stale run_008/run_009 dirs
- Ruff config moved to [tool.ruff.lint] (silences deprecation warning)
- Pre-commit: bump revs (ruff v0.14.13, mypy v1.18.1, hooks v6.0.0);
  drop redundant isort hook (ruff I already sorts imports)
- mypy: add ignore_missing_imports for optional deps; types-PyYAML in
  dev extras; python_version=3.12 for numpy 2.5 stubs compat; CI step
- Doctor: suppress paddle ccache UserWarning noise during check
- CHANGELOG dated for 0.2.0; README minimal-style rewrite with
  collapsible details sections
- Bump version to 0.2.0

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ocr-pipeline"
-version = "0.1.0"
+version = "0.2.0"
 description = "OCR pipeline for life science screenshots - RAG ready"
 readme = "README.md"
 requires-python = ">=3.11"
@@ -42,6 +42,7 @@ dev = [
     "pytest-mock>=3.12.0",
     "ruff>=0.2.0",
     "mypy>=1.8.0",
+    "types-PyYAML>=6.0.12",
     "pre-commit>=3.6.0",
 ]
 paddle = [
@@ -93,6 +94,7 @@ dev-dependencies = [
     "pytest-mock>=3.12.0",
     "ruff>=0.2.0",
     "mypy>=1.8.0",
+    "types-PyYAML>=6.0.12",
     "pre-commit>=3.6.0",
 ]
 override-dependencies = ["lxml>=5.3.0"]
@@ -100,6 +102,8 @@ override-dependencies = ["lxml>=5.3.0"]
 [tool.ruff]
 line-length = 100
 target-version = "py311"
+
+[tool.ruff.lint]
 select = [
     "E",   # pycodestyle
     "F",   # pyflakes
@@ -125,17 +129,39 @@ skip-magic-trailing-comma = false
 line-ending = "lf"
 
 [tool.mypy]
-python_version = "3.11"
+python_version = "3.12"  # numpy 2.5 stubs need >= 3.12 type-syntax
 warn_return_any = true
 warn_unused_configs = true
-disallow_untyped_defs = true
-disallow_incomplete_defs = true
 check_untyped_defs = true
 no_implicit_optional = true
 strict_equality = true
 show_error_codes = true
 pretty = true
 
+# Optional extras: stubs unavailable or packages absent in slim installs.
+[[tool.mypy.overrides]]
+module = [
+    "paddleocr",
+    "paddle",
+    "torch",
+    "torchvision",
+    "transformers",
+    "layoutparser",
+    "spacy",
+    "pytesseract",
+    "cv2",
+    "PIL",
+    "txtai",
+    "txtai.embeddings",
+    "litellm",
+    "neo4j",
+    "networkx",
+    "watchdog",
+    "langchain.*",
+    "nbformat",
+]
+ignore_missing_imports = true
+
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 python_files = ["test_*.py"]