perf(ci): Cut Docker smoke-test time on large runners (#588)

## What
Two changes to the Docker smoke-test job:

- **Disk cleanup runs only when space is short.** Cleanup is skipped
when both `/` and Docker's storage filesystem have at least 30 GiB free.
Smaller runners still clean up as before.
- **The generic PDF and `.gz` format checks use the one-page PDF**
instead of the full paper. The `.gz` case compresses that PDF in
pytest's temp directory, so the unused `layout-parser-paper.pdf.gz`
fixture is removed.

```
Before: cleanup 183s (with 371 GiB free) · .pdf case 175s · .pdf.gz case 191s
After:  cleanup skipped                  · both cases process a one-page PDF
```

## Why
These checks only confirm that each file type is accepted and parsed.
The full paper is still processed by the table-extraction, strategy, and
parallel-mode smoke tests, so the generic format cases don't need it.
All 37 cases and their explicit-MIME, inferred-MIME, and CSV assertions
remain.

## Validation
actionlint and Ruff pass, and all 37 smoke tests collect. Both changed
cases were run against a controlled response, and the gzip decompresses
byte-for-byte to the source PDF. Both disk branches (enough space and
low space) were tested with mocked commands. Full timing comes from this
PR's CI
([baseline](https://github.com/Unstructured-IO/unstructured-api/actions/runs/36955856924)).

Author: CyMule

.github/workflows/ci.yml

@@ -102,26 +102,23 @@ jobs:
         cache-suffix: smoke
     - name: Set up Python ${{ env.PYTHON_VERSION }}
       run: uv python install ${{ env.PYTHON_VERSION }}
-    - name: Free up disk space
+    - name: Free up disk space when needed
       run: |
-        # Clear some space (https://github.com/actions/runner-images/issues/2840)
-        echo "Disk usage before cleanup:"
-        df -h
-
-        # Remove unnecessary pre-installed software
-        sudo rm -rf /usr/share/dotnet
-        sudo rm -rf /opt/ghc
-        sudo rm -rf /usr/local/share/boost
-        sudo rm -rf /usr/local/lib/android
-        sudo rm -rf /opt/hostedtoolcache/CodeQL
-        sudo rm -rf /usr/local/.ghcup
-        sudo rm -rf /usr/share/swift
-
-        # Clean up docker to ensure we start fresh
+        # Large runners normally have hundreds of GiB free. Keep the fallback for
+        # smaller runners, but avoid spending minutes deleting unused toolchains.
+        # Docker images and the host toolchains/venv may live on different
+        # filesystems, so both must have room before cleanup is skipped.
+        docker_root=$(docker info --format '{{.DockerRootDir}}')
+        available_kib=$(df -Pk / "$docker_root" | awk 'NR > 1 && (min == "" || $4 < min) {min = $4} END {print min}')
+        if (( available_kib >= 30 * 1024 * 1024 )); then
+          echo "Skipping cleanup: $((available_kib / 1024 / 1024)) GiB free on / and Docker storage"
+          exit 0
+        fi
+        df -h / "$docker_root"
+        sudo rm -rf /usr/share/dotnet /opt/ghc /usr/local/share/boost \
+          /usr/local/lib/android /opt/hostedtoolcache/CodeQL /usr/local/.ghcup /usr/share/swift
         docker system prune -af --volumes
-
-        echo "Disk usage after cleanup:"
-        df -h
+        df -h / "$docker_root"
     - name: Install smoke test dependencies
       run: uv sync --only-group smoke --locked
     - name: Build Dockerfile

sample-docs/layout-parser-paper.pdf.gz

@@ -67,7 +67,7 @@ def send_document(
         (".md", "README.md", "text/markdown"),
         (".msg", "fake-email.msg", "application/x-ole-storage"),
         (".odt", "fake.odt", "application/vnd.oasis.opendocument.text"),
-        (".pdf", "layout-parser-paper.pdf", "application/pdf"),
+        (".pdf", "layout-parser-paper-fast.pdf", "application/pdf"),
         (".png", "english-and-korean.png", "image/png"),
         (".ppt", "fake-power-point.ppt", "application/vnd.ms-powerpoint"),
         (
@@ -89,12 +89,12 @@ def send_document(
         (".json", "spring-weather.html.json", "application/json"),
         (
             ".gz",
-            "layout-parser-paper.pdf.gz",
+            "layout-parser-paper-fast.pdf",
             "application/gzip",
         ),
     ],
 )
-def test_happy_path_all_types(extension, example_filename: str, content_type: str):
+def test_happy_path_all_types(extension, example_filename: str, content_type: str, tmp_path: Path):
     """
     For the files in sample-docs, verify that we get a 200
     and some structured response
@@ -113,6 +113,12 @@ def test_happy_path_all_types(extension, example_filename: str, content_type: st
         pytest.skip("emulated hardware")
 
     test_file = str(Path("sample-docs") / example_filename)
+    if extension == ".gz":
+        # Gzip the one-page PDF at test time; full-document inference is covered
+        # by the table and strategy tests.
+        gzipped_file = tmp_path / f"{example_filename}.gz"
+        gzip_file(test_file, str(gzipped_file))
+        test_file = str(gzipped_file)
 
     # Verify we can send with explicit content type
     response = send_document(filenames=[test_file], content_type=content_type)