reextract: process in batches of 100 to bound subprocess memory

One Python process for 2015 activities exhausts all RAM + swap on a
cheap VPS. Split into sequential batches of 100: each subprocess handles
100 activities and exits, returning all memory to the OS before the
next batch starts. The server chains batches in the SSE event_stream
and triggers a single rebuild when all batches complete.

bincio/reextract_cmd.py

@@ -33,7 +33,9 @@ _GC_EVERY = 50  # call gc.collect() + malloc_trim every N activities
 @click.option("--data-dir", required=True, type=click.Path(), help="BAS data directory")
 @click.option("--handle", required=True, help="User handle to re-extract for")
 @click.option("--force", is_flag=True, default=False, help="Re-extract even if activity JSON already exists")
-def reextract_originals(data_dir: str, handle: str, force: bool) -> None:
+@click.option("--offset", default=0, type=int, help="Skip first N originals (for batch processing)")
+@click.option("--limit", default=0, type=int, help="Process at most N originals then stop (0 = all)")
+def reextract_originals(data_dir: str, handle: str, force: bool, offset: int, limit: int) -> None:
     """Re-extract activities from stored Strava originals (originals/strava/*.json).
 
     Prints one JSON object per line to stdout for streaming progress:
@@ -57,13 +59,22 @@ def reextract_originals(data_dir: str, handle: str, force: bool) -> None:
         _emit({"type": "error", "message": f"No Strava originals directory at {originals_dir}"})
         sys.exit(1)
 
-    original_files = sorted(originals_dir.glob("*.json"))
-    total = len(original_files)
-    if total == 0:
+    all_files = sorted(originals_dir.glob("*.json"))
+    if not all_files:
         _emit({"type": "error", "message": "No Strava originals found"})
         sys.exit(1)
 
-    _emit({"type": "status", "message": f"Found {total} originals, starting extraction…"})
+    # Apply offset/limit for batch processing
+    batch = all_files[offset:] if not limit else all_files[offset: offset + limit]
+    total_all = len(all_files)
+    total = len(batch)
+    original_files = batch
+
+    _emit({"type": "status", "message": (
+        f"Batch {offset + 1}–{offset + total} of {total_all}, starting extraction…"
+        if offset or limit else
+        f"Found {total_all} originals, starting extraction…"
+    )})
 
     # Load existing index to get owner info and existing summaries
     index_path = user_dir / "index.json"