fix: search for failing file in read context (SeaweedInputStream)

The issue: We're not finding the correct file because:
1. Error mentions: test-spark/employees/part-00000-xxx.parquet
2. But we downloaded chunk from employees_window (different file!)

The problem:
- File is already written when error occurs
- Error happens during READ, not write
- Need to find when SeaweedInputStream opens this file for reading

New approach:
1. Extract filename from EOF error message
2. Search for 'new path:' + filename (when file is opened for read)
3. Get chunk info from the entry details logged at that point
4. Download the ACTUAL failing chunk

This should finally get us the right file with the 78-byte issue!

.github/workflows/spark-integration-tests.yml

@@ -139,12 +139,34 @@ jobs:
               FULL_LOG=$(docker compose logs spark-tests 2>&1)
               
               # Extract the failing filename from the EOF error message
-              FAILING_FILE=$(echo "$FULL_LOG" | grep "EOFException" | grep -oP 'test-spark/employees/\K[^"]+\.parquet' | head -1)
+              # The error message format: "...seaweedfs://seaweedfs-filer:8888/test-spark/employees/part-xxx.parquet..."
+              FAILING_FILE=$(echo "$FULL_LOG" | grep -B 5 "EOFException.*78 bytes" | grep "seaweedfs://" | grep -oP 'part-[a-f0-9-]+\.c000\.snappy\.parquet' | head -1)
               echo "Failing file: $FAILING_FILE"
               
+              if [ -z "$FAILING_FILE" ]; then
+                echo "ERROR: Could not extract failing filename from error message"
+                echo "Searching for error message pattern..."
+                echo "$FULL_LOG" | grep -A 2 "EOFException.*78 bytes" | head -20
+                break
+              fi
+              
               # Now find the chunk info for THIS SPECIFIC FILE
-              # Search backwards from the filename to find its chunk info
-              CHUNK_ID=$(echo "$FULL_LOG" | grep -B 200 "$FAILING_FILE" | grep 'chunks {' -A 10 | grep '  file_id: "' | tail -1 | grep -oP '"\K[^"]+')
+              # The file is being READ when the error occurs, so look for SeaweedInputStream opening it
+              echo "Searching logs for when $FAILING_FILE was opened for reading..."
+              
+              # Find all instances where this file is mentioned and get nearby chunk info
+              # Strategy: Search for the filename, then look for "chunks {" blocks near it
+              CHUNK_CONTEXT=$(echo "$FULL_LOG" | grep -A 100 "new path:.*$FAILING_FILE")
+              
+              if [ -n "$CHUNK_CONTEXT" ]; then
+                echo "Found read context for file"
+                CHUNK_ID=$(echo "$CHUNK_CONTEXT" | head -30 | grep '  file_id: "' | head -1 | grep -oP '"\K[^"]+')
+              else
+                echo "No read context, trying write context..."
+                # Maybe it's in the write logs
+                CHUNK_CONTEXT=$(echo "$FULL_LOG" | grep -B 50 -A 20 "$FAILING_FILE")
+                CHUNK_ID=$(echo "$CHUNK_CONTEXT" | grep '  file_id: "' | head -1 | grep -oP '"\K[^"]+')
+              fi
               echo "Found chunk ID: $CHUNK_ID"
               
               if [ -n "$CHUNK_ID" ]; then