docker: Use conda to install dependencies; add conda environment

This results in both a shorter Dockerfile and a smaller final image.
The conda environment matches the existing conda packaging in devtools/.

conda-env.yml

@@ -0,0 +1,29 @@
+# Usage:
+#   conda env create -f conda-env.yaml
+# or:
+#   conda env update -v -n [base] -f conda-env.yml [--prune]
+# This does not install CNVkit itself. Use pip to install either a stable release from
+# PyPI (pip install cnvkit), or this local repo (pip install -e .).
+name: cnvkit
+channels:
+    - conda-forge
+    - bioconda
+    - defaults
+dependencies:
+    - python>=3.8
+    - atlas # [not osx]
+    - bioconductor-dnacopy
+    - biopython >=1.80
+    - matplotlib >=3.5.2
+    - numpy >=1.24.2
+    - pandas >=1.5.3
+    - pomegranate >=0.14.8, <=0.14.9
+    - pyfaidx >=0.7.1
+    - pysam >=0.20.0
+    - pytest
+    - reportlab >=3.6.12
+    - ruff
+    - scikit-learn >=1.1.0
+    - scipy >=1.10.1
+    - tox
+

docker/Dockerfile

@@ -1,18 +1,11 @@
-FROM ubuntu:rolling
+FROM continuumio/miniconda3:23.10.0-1
 MAINTAINER Eric Talevich <me+code@etal.mozmail.com>
 
-ENV DEBIAN_FRONTEND noninteractive
-RUN apt-get update && apt-get install -y \
-    liblzma-dev \
-    python3-dev \
-    python3-pip \
-    r-bioc-dnacopy \
-    zlib1g-dev \
-    && rm -rf /var/lib/apt/lists/*
-
-RUN pip3 install --upgrade pip --break-system-packages && \
-    pip3 install cnvkit==0.9.10 --break-system-packages && \
-    pip3 cache purge
+# Install directly into 'base' conda environment
+COPY conda-env.yml ./conda-env.yml
+RUN conda env update -v -n base -f conda-env.yml
+RUN conda clean --all --verbose
+RUN pip3 install cnvkit==0.9.10 --no-cache
 # Let matplotlib build its font cache
 RUN cnvkit.py version