Add --metadata-only mode; harden resume and idempotency

- Add --metadata-only flag: fetches scan detail pages, writes
  metadata.json + scans.csv rows, skips all image downloads.
  Re-runs skip scans whose metadata.json already exists.
- Atomic progress.json saves (temp-file rename).
- Heal-on-resume: tiles on disk but not in progress are silently
  re-marked before building the pending list.
- scans.csv dedup: skip row if mosaic URL already in progress.
- Rename mosaic_downloaded -> mosaic_on_disk (reflects disk state).
- --recheck now checks mosaics as well as tiles.
- RunStats dataclass replaces raw int return; richer run summary.
- Fix argparse allow_abbrev reverted; fix --scan-id + --metadata-only
  glob fallback when scan_time is absent.
- Add .venv/ to .gitignore.
- README: fix typo, update worker counts, document all new behaviour.

spruce/cli.py

@@ -10,7 +10,7 @@ from pathlib import Path
 
 import yaml
 
-from spruce.orchestrator import scrape_machine
+from spruce.orchestrator import scrape_machine, RunStats
 from spruce.parsers import parse_machine_option
 from spruce.progress import ProgressTracker, CsvWriter
 from spruce.recheck import recheck_archive, recheck_tile_files
@@ -75,6 +75,15 @@ def parse_args() -> argparse.Namespace:
         action="store_true",
         help="Download mosaics only; skip individual tiles",
     )
+    p.add_argument(
+        "--metadata-only",
+        action="store_true",
+        help=(
+            "Fetch scan parameters only; write metadata.json and scans.csv "
+            "rows but skip mosaics and tiles. Very fast — suitable for "
+            "inventorying all scans across all machines."
+        ),
+    )
     p.add_argument(
         "--dry-run",
         action="store_true",
@@ -220,7 +229,11 @@ def main() -> None:
         len(machines),
         ", ".join(m["label"] for m in machines),
     )
-    if args.mosaic_only:
+    if args.mosaic_only and args.metadata_only:
+        sys.exit("--mosaic-only and --metadata-only are mutually exclusive.")
+    if args.metadata_only:
+        log.info("Mode: metadata only (mosaics and tiles skipped)")
+    elif args.mosaic_only:
         log.info("Mode: mosaics only (individual tiles skipped)")
     if args.dry_run:
         log.info("Mode: dry-run (no files will be written)")
@@ -230,10 +243,10 @@ def main() -> None:
     tiles_csv = CsvWriter(output_dir / TILES_CSV_FILENAME, TILES_CSV_FIELDS)
     scans_csv = CsvWriter(output_dir / SCANS_CSV_FILENAME, SCANS_CSV_FIELDS)
 
-    total = 0
+    totals = RunStats()
     try:
         for machine in machines:
-            count = scrape_machine(
+            stats = scrape_machine(
                 machine=machine,
                 config=config,
                 output_dir=output_dir,
@@ -242,18 +255,62 @@ def main() -> None:
                 scans_csv=scans_csv,
                 dry_run=args.dry_run,
                 mosaic_only=args.mosaic_only,
+                metadata_only=args.metadata_only,
                 scan_id_filter=args.scan_id,
             )
-            total += count
+            totals.merge(stats)
     finally:
         tiles_csv.close()
         scans_csv.close()
         progress.save()
 
-    if args.dry_run:
-        log.info("Dry run complete.")
+    _print_summary(
+        totals=totals,
+        machines=machines,
+        output_dir=output_dir,
+        dry_run=args.dry_run,
+        metadata_only=args.metadata_only,
+        mosaic_only=args.mosaic_only,
+    )
+
+
+def _print_summary(
+    totals: RunStats,
+    machines: list[dict],
+    output_dir: Path,
+    dry_run: bool,
+    metadata_only: bool,
+    mosaic_only: bool,
+) -> None:
+    W = 46
+    sep = "─" * W
+
+    def row(label: str, value: str, note: str = "") -> str:
+        note_str = f"  ({note})" if note else ""
+        return f"  {label:<22}{value}{note_str}"
+
+    log.info(sep)
+    if dry_run:
+        log.info("  Dry run complete — no files written.")
     else:
-        log.info("Done. Total files downloaded: %d", total)
-        log.info("Tiles CSV : %s", output_dir / TILES_CSV_FILENAME)
-        log.info("Scans CSV : %s", output_dir / SCANS_CSV_FILENAME)
-        log.info("Progress  : %s", output_dir / PROGRESS_FILENAME)
+        log.info("  Run complete")
+        log.info(sep)
+        log.info(row("Machines:", str(len(machines))))
+        log.info(
+            row("Scans fetched:", str(totals.scans_fetched),
+                f"{totals.scans_skipped} already cached, "
+                f"{totals.scans_failed} failed"
+                if totals.scans_skipped or totals.scans_failed else "")
+        )
+        if not metadata_only:
+            log.info(row("Mosaics downloaded:", str(totals.mosaics_downloaded)))
+        if not metadata_only and not mosaic_only:
+            log.info(row("Tiles downloaded:", str(totals.tiles_downloaded)))
+        if metadata_only:
+            log.info(row("Metadata written:", str(totals.metadata_written), "new JSON files"))
+    log.info(sep)
+    log.info(row("Scans CSV:", str(output_dir / SCANS_CSV_FILENAME)))
+    if not metadata_only:
+        log.info(row("Tiles CSV:", str(output_dir / TILES_CSV_FILENAME)))
+    log.info(row("Progress:", str(output_dir / PROGRESS_FILENAME)))
+    log.info(sep)