scripts/lambda-bootstrap.sh — drop the cd-into-repo / before launching trainer

The previous version did `(cd repo && "${cmd[@]}")` to "cd into repo
for module imports." But PYTHONPATH was already set to $PWD/repo at
the top of the script — so the cd was redundant for imports AND
broke relative paths: the trainer expects to find
data/processed/validation_v1.parquet from $HOME/cis490, not from
$HOME/cis490/repo/.

Symptom: every training job failed immediately with
  FileNotFoundError: data/processed/validation_v1.parquet

Drop the cd; PYTHONPATH already does the import work.

Found while running on the A100 today; trainer relaunched manually
in-place via a stand-in bootstrap2.sh; this commit makes the next
bundle clean.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

scripts/lambda-bootstrap.sh

@@ -175,10 +175,10 @@ for entry in "${JOBS[@]}"; do
         cmd+=("${extra_args[@]}")
     fi
 
-    # Launch in background. Each subshell cd's into repo/ for module
-    # imports; output redirected to per-job log; trainer + torch handle
-    # multi-process CUDA OK on a single A100.
-    (cd repo && "${cmd[@]}") > "$log" 2>&1 &
+    # Launch in background. PYTHONPATH is set to $PWD/repo at the top
+    # of this script so we DO NOT cd into repo/ — relative paths to
+    # data/processed/* must resolve from $HOME/cis490, not from repo/.
+    "${cmd[@]}" > "$log" 2>&1 &
     pid=$!
     PIDS+=("$pid")
     PID_TO_LABEL[$pid]="$job_label"