Update Dockerfile

Dockerfile

@@ -77,8 +77,10 @@ ENV DEBIAN_FRONTEND=noninteractive \
     # Force CPU-only for PyTorch/transformers
     CUDA_VISIBLE_DEVICES= \
     PYTORCH_ENABLE_MPS=0 \
+    XDG_CACHE_HOME=/app/.cache \
     PYTORCH_NO_CUDA=1
 
+
 # Copy only what's needed (avoid duplicate COPY . /app)
 COPY requirements.txt ./requirements.txt
 COPY app.py ./app.py
@@ -89,6 +91,8 @@ COPY generated_projects/ ./generated_projects/
 
 # Install system dependencies (minimal for CPU inference + common tools)
 RUN apt-get update && apt-get install -y --no-install-recommends \
+    fontconfig \
+    fonts-dejavu-core \
     build-essential \
     curl \
     ca-certificates \
@@ -101,6 +105,11 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     poppler-utils \
     && apt-get clean && rm -rf /var/lib/apt/lists/*
 
+RUN python -c "import nltk; \
+    nltk.download('punkt', download_dir='/app/nltk_data'); \
+    nltk.download('averaged_perceptron_tagger', download_dir='/app/nltk_data'); \
+    nltk.download('wordnet', download_dir='/app/nltk_data')"
+
 # Python deps
 RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt
 
@@ -110,6 +119,13 @@ RUN mkdir -p /app/nltk_data /app/.config/matplotlib \
  && mkdir -p /app/outputs/DETECTED_IMAGE /app/outputs/SCANNED_IMAGE /app/outputs/EXTRACTED_JSON \
  && chown -R root:root /app \
  && chmod -R 755 /app/cache /app/data /app/logs /app/outputs
+ 
+ RUN mkdir -p /app/.config/matplotlib \
+    /app/cache /app/nltk_data /nltk_data \
+    /app/.cache/fontconfig /root/.cache/fontconfig \
+    && chmod -R 777 /app/.config/matplotlib /app/.cache /app/cache /app/nltk_data /nltk_data /root/.cache/fontconfig
+
+RUN fc-cache -f -v || true
 
 # Set Flask env
 ENV FLASK_APP=app.py \