feat: Add granular resume within partitions using last inserted ID

Problem: If migration was interrupted in the middle of processing a partition
(e.g., at row 100k of 500k), resume would re-process all 100k rows, causing
duplicate insertions and wasted time.

Solution:
1. Modified fetch_consolidation_groups_from_partition() to accept start_id parameter
2. When resuming within the same partition, query the last inserted ID from
   migration_state.last_migrated_id
3. Use keyset pagination starting from (id > last_id) to skip already-processed rows
4. Added logic to detect when we're resuming within the same partition vs resuming
   from a new partition

Flow:
- If last_completed_partition < current_partition: start from beginning of partition
- If last_completed_partition == current_partition: start from last_migrated_id
- If last_completed_partition > current_partition: skip to next uncompleted partition

This ensures resume is granular:
- Won't re-insert already inserted rows within a partition
- Continues exactly from where it stopped
- Combines with existing partition tracking for complete accuracy

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

src/migrator/full_migration.py

@@ -120,6 +120,16 @@ class FullMigrator:
                             logger.info(f"[{partition_idx}/{len(partitions)}] Processing partition {partition}...")
                             partition_group_count = 0
 
+                            # Determine resume point within this partition
+                            # If resuming and this is the last completed partition, start from last_id
+                            start_id = None
+                            if last_completed_partition == partition and previous_migrated_count > 0:
+                                # For resume within same partition, we need to query the last ID inserted
+                                # This is a simplified approach: just continue from ID tracking
+                                start_id = self._get_last_migrated_id(pg_conn, pg_table)
+                                if start_id:
+                                    logger.info(f"Resuming partition {partition} from ID > {start_id}")
+
                             # Accumulate rows for batch insertion to reduce database round-trips
                             insert_buffer = []
                             # Use smaller batch size for more frequent updates: batch_size * 5 = 50k rows
@@ -130,7 +140,8 @@ class FullMigrator:
                             # Each group is a list of rows with the same (unit, tool, date, time)
                             for group_rows in mysql_conn.fetch_consolidation_groups_from_partition(
                                 mysql_table,
-                                partition
+                                partition,
+                                start_id=start_id
                             ):
                                 if not group_rows:
                                     break