feat(docker): use multi-stage Docker builds for smaller images (#1614)

# Description

Currently, the production Docker images are very large, sitting at 4.17
GB for the frontend image, and 3.49 GB for backend images. This change
adds multi-stage builds, to optimize the image sizes, which results in
the following improvements:

- frontend image size: 4.17 GB -> 1.64 GB
- backend image size: 3.49 GB -> 1.71 GB

I hope this is appropriate as there is no open issue for this that I
know of.
I implemented this change and tested it locally, and would be glad to
discuss this and open an issue if necessary.

## Checklist before requesting a review

Please delete options that are not relevant.

- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my code
- [x] I have commented hard-to-understand areas
- [x] New and existing unit tests pass locally with my changes

## Screenshots (if appropriate):
Image sizes before:
<img width="1416" alt="image"
src="https://github.com/StanGirard/quivr/assets/8296549/fcbb020f-8165-4549-ae30-823318691ec6">

Image sizes after:
<img width="1416" alt="image"
src="https://github.com/StanGirard/quivr/assets/8296549/d3f43c78-be26-4c38-9d23-9c1b0e9e37f2">

backend/Dockerfile

@@ -1,10 +1,10 @@
-# Using a slim version for a smaller base image
-FROM python:3.11-slim-bullseye
+# First stage: Build environment
+FROM python:3.11-slim-bullseye as builder
 
 ARG DEV_MODE
 ENV DEV_MODE=$DEV_MODE
 
-# Install GEOS library, Rust, and other dependencies, then clean up
+# Install GEOS library, Rust, and other build-time dependencies
 RUN apt-get update && apt-get install -y \
     libgeos-dev \
     libcurl4-openssl-dev \
@@ -20,19 +20,43 @@ RUN apt-get update && apt-get install -y \
 # Add Rust binaries to the PATH
 ENV PATH="/root/.cargo/bin:${PATH}"
 
+# Create a virtual environment and activate it
+RUN python -m venv /venv
+ENV PATH="/venv/bin:$PATH"
+
 WORKDIR /code
 
 # Copy just the requirements first
 COPY ./requirements.txt .
 
-# Upgrade pip
-RUN pip install --upgrade pip
-
-# Increase timeout to wait for the new installation
-RUN pip install --no-cache-dir -r requirements.txt --timeout 200
+# Upgrade pip and install requirements
+RUN pip install --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt --timeout 200
 
+# Install development tools if in DEV_MODE
 RUN if [ "$DEV_MODE" = "true" ]; then pip install --no-cache debugpy --timeout 200; fi
 
+
+
+FROM python:3.11-slim-bullseye as runtime
+
+ARG DEV_MODE
+ENV DEV_MODE=$DEV_MODE
+
+# Create and activate the virtual environment
+ENV PATH="/venv/bin:$PATH"
+
+# Install runtime dependencies
+RUN apt-get update && apt-get install -y \
+    pandoc \
+    binutils && \
+    rm -rf /var/lib/apt/lists/* && apt-get clean
+
+WORKDIR /code
+
+# Copy the virtual environment from the builder stage
+COPY --from=builder /venv /venv
+
 # Copy the rest of the application
 COPY . .
 

frontend/Dockerfile

@@ -1,31 +1,49 @@
-FROM node:18.13.0-alpine
+# First stage: Build environment
+FROM node:18.13.0-alpine as builder
+
 # Install Python and essential build tools
 RUN apk add --update --no-cache python3 make g++ && ln -sf python3 /usr/bin/python
 RUN python3 -m ensurepip
 RUN pip3 install --no-cache --upgrade pip setuptools
 
-# Create the directory on the node image 
-# where our Next.js app will live
-RUN mkdir -p /app
+# Create a Python virtual environment
+RUN python3 -m venv /venv
+ENV PATH="/venv/bin:$PATH"
 
-# Set /app as the working directory
+# Create the directory where our app will live
+RUN mkdir -p /app
 WORKDIR /app
 
-# Copy package.json and yarn.lock
-# to the /app working directory
+# Copy package.json and yarn.lock to the working directory
 COPY package*.json yarn.lock ./
 
-# Install dependencies in /app
+# Install Node.js dependencies
 RUN yarn install --network-timeout 1000000
 
-# Copy the rest of our Next.js folder into /app
+# Copy the rest of the Next.js folder into /app
 COPY . .
 
 # Build the Next.js application
 RUN yarn build
 
+# Second stage: Runtime environment
+FROM node:18.13.0-alpine
+
+# Copy the virtual environment from the builder stage
+COPY --from=builder /venv /venv
+ENV PATH="/venv/bin:$PATH"
+
+# Set the working directory
+WORKDIR /app
+
+# Copy built assets from the builder stage
+COPY --from=builder /app/.next ./.next
+COPY --from=builder /app/node_modules ./node_modules
+COPY --from=builder /app/package.json ./package.json
+COPY --from=builder /app/public ./public
+
 # Ensure port 3000 is accessible to our system
 EXPOSE 3000
 
-# Run yarn start, as we would via the command line 
+# Run yarn start, as we would via the command line
 CMD ["yarn", "start"]