Try using mambaforge instead of conda

I don't think this will help

deployments/datahub/images/default/install-miniforge.bash

@@ -1,12 +1,12 @@
 #!/bin/bash
-# This downloads and installs a pinned version of miniconda
+# This downloads and installs a pinned version of mambaforge
 set -ex
 
 cd $(dirname $0)
-MINIFORGE_VERSION=4.10.3-4
+MAMBAFORGE_VERSION=4.10.3-7
 
-URL="https://github.com/conda-forge/miniforge/releases/download/${MINIFORGE_VERSION}/Miniforge3-${MINIFORGE_VERSION}-Linux-x86_64.sh"
-INSTALLER_PATH=/tmp/miniforge-installer.sh
+URL="https://github.com/conda-forge/miniforge/releases/download/${MAMBAFORGE_VERSION}/Mambaforge-${MAMBAFORGE_VERSION}-Linux-x86_64.sh"
+INSTALLER_PATH=/tmp/mambaforge-installer.sh
 
 # make sure we don't do anything funky with user's $HOME
 # since this is run as root
@@ -34,7 +34,7 @@ conda clean --all -f -y
 # Remove the big installer so we don't increase docker image size too much
 rm ${INSTALLER_PATH}
 
-# Remove the pip cache created as part of installing miniconda
+# Remove the pip cache created as part of installing mambaforge
 rm -rf /root/.cache
 
 chown -R $NB_USER:$NB_USER ${CONDA_DIR}