feat(ml): export clip models to ONNX and host models on Hugging Face (#4700)

* export clip models

* export to hf

refactored export code

* export mclip, general refactoring

cleanup

* updated conda deps

* do transforms with pillow and numpy, add tokenization config to export, general refactoring

* moved conda dockerfile, re-added poetry

* minor fixes

* updated link

* updated tests

* removed `requirements.txt` from workflow

* fixed mimalloc path

* removed torchvision

* cleaner np typing

* review suggestions

* update default model name

* update test

machine-learning/export/Dockerfile

@@ -0,0 +1,21 @@
+FROM mambaorg/micromamba:bookworm-slim as builder
+
+ENV NODE_ENV=production \
+  TRANSFORMERS_CACHE=/cache \
+  PYTHONDONTWRITEBYTECODE=1 \
+  PYTHONUNBUFFERED=1 \
+  PATH="/opt/venv/bin:$PATH" \
+  PYTHONPATH=/usr/src
+
+COPY --chown=$MAMBA_USER:$MAMBA_USER conda-lock.yml /tmp/conda-lock.yml
+RUN micromamba install -y -n base -f /tmp/conda-lock.yml && \
+    micromamba remove -y -n base cxx-compiler && \
+    micromamba clean --all --yes
+
+WORKDIR /usr/src/app
+
+COPY --chown=$MAMBA_USER:$MAMBA_USER start.sh .
+COPY --chown=$MAMBA_USER:$MAMBA_USER app .
+
+ENTRYPOINT ["/usr/local/bin/_entrypoint.sh"]
+CMD ["./start.sh"]