initial whl workflow (#57)

SUMMARY:
* add helper script for building in python venv
* add GHA action for building whl file
* add helper script for converting python version to whl style
* add callable workflow to use for constructing whl files

TEST PLAN:
this is a seed PR ... testing will commence once these changes are
merged

---------

Co-authored-by: andy-neuma <andy@neuralmagic.com>

.github/actions/nm-build-vllm-whl/action.yml

@@ -0,0 +1,33 @@
+name: build neuralmagic-vllm whl
+description: 'build whl file of neuralmagic-vllm'
+inputs:
+  python:
+    description: 'python version, e.g. 3.10.12'
+    required: true
+  venv:
+    description: 'name for python virtual environment'
+    required: true
+outputs:
+  status:
+    description: "final build status from 'pip3 wheel dist .'"
+    value: ${{ steps.whl.outputs.status }}
+  whl:
+    description: 'basename for generated whl'
+    value: ${{ steps.build_whl.outputs.whl }}
+runs:
+  using: composite
+  steps:
+  - id: whl
+    run: |
+      COMMIT=${{ github.sha }}
+      VENV="${{ inputs.venv }}-${COMMIT:0:7}"
+      source $(pyenv root)/versions/${{ inputs.python }}/envs/${VENV}/bin/activate
+      SUCCESS=0
+      pip3 wheel --no-deps -w dist . || SUCCESS=$?
+      echo "status=${SUCCESS}" >> "$GITHUB_OUTPUT"
+      BASE=$(./.github/scripts/convert-version ${{ inputs.python }})
+      WHL_FILEPATH=$(find dist -iname "*nm_vllm*${BASE}*.whl")
+      WHL=$(basename ${WHL_FILEPATH})
+      echo "whl=${WHL}" >> "$GITHUB_OUTPUT"
+      exit ${SUCCESS}
+    shell: bash

.github/scripts/build

@@ -0,0 +1,60 @@
+#!/usr/bin/env bash
+
+usage() {
+    echo``
+    echo "build 'neuralmagic-vllm' in specified python and venv"
+    echo
+    echo "usage: ${0} <options>"
+    echo
+    echo "  -a    - pypi server address"
+    echo "  -p    - python version"
+    echo "  -v    - name for virtualenv"
+    echo "  -h    - this list of options"
+    echo
+}
+
+PYPI_IP=
+PYTHON=
+VENV=
+
+while getopts "ha:p:v:" OPT; do
+    case "${OPT}" in
+        h)
+            usage
+            exit 1
+            ;;
+        a)
+            PYPI_IP="${OPTARG}"
+            ;;
+        p)
+            PYTHON="${OPTARG}"
+            ;;
+        v)
+            VENV="${OPTARG}"
+            ;;
+    esac
+done
+
+# check if variables are valid
+if [ -z "${PYPI_IP}" ]; then
+    echo "please provide 'pypi' server address"
+    usage
+    exit 1
+fi
+
+if [ -z "${PYTHON}" ]; then
+    echo "please provide python version, e.g. 3.10.12"
+    usage
+    exit 1
+fi
+
+if [ -z "${VENV}" ]; then
+    echo "please provide python virutalenv name, e.g. TEST"
+    usage
+    exit 1
+fi
+
+source $(pyenv root)/versions/${PYTHON}/envs/${VENV}/bin/activate
+pip3 install --index-url http://${PYPI_IP}:8080/ --trusted-host ${PYPI_IP} magic-wand
+pip3 install -r requirements.txt
+pip3 install -e .

.github/scripts/convert-version

@@ -0,0 +1,20 @@
+#!/bin/bash -e
+
+# convert python version to whl style
+INPUT=${1}
+BASE=
+
+if [[ "${INPUT}" == *"3.11."* ]]; then
+    BASE="cp311-cp311";
+elif [[ "${INPUT}" == *"3.10."* ]]; then
+    BASE="cp310-cp310";
+elif [[ "${INPUT}" == *"3.9."* ]]; then
+    BASE="cp39-cp39";
+elif [[ "${INPUT}" == *"3.8."* ]]; then
+    BASE="cp38-cp38";
+else
+    echo "unsupported or unspecified python version '${1}'"
+    exit 1
+fi
+
+echo "${BASE}"

.github/workflows/build-whl.yml

@@ -0,0 +1,127 @@
+name: build-whl
+on:
+  # makes workflow reusable
+  workflow_call:
+    inputs:
+      build_label:
+        description: "requested runner label (specifies instance)"
+        type: string
+        required: true
+      timeout:
+        description: "time limit for run in minutes "
+        type: string
+        required: true
+      gitref:
+        description: "git commit hash or branch name"
+        type: string
+        required: true
+      Gi_per_thread:
+        description: 'requested GiB to reserve per thread'
+        type: string
+        required: true
+      python:
+        description: "python version, e.g. 3.10.12"
+        type: string
+        required: true
+
+  # makes workflow manually callable
+  workflow_dispatch:
+    inputs:
+      build_label:
+        description: "requested runner label (specifies instance)"
+        type: string
+        required: true
+      timeout:
+        description: "time limit for run in minutes "
+        type: string
+        required: true
+      gitref:
+        description: "git commit hash or branch name"
+        type: string
+        required: true
+      Gi_per_thread:
+        description: 'requested GiB to reserve per thread'
+        type: string
+        required: true
+      python:
+        description: "python version, e.g. 3.10.12"
+        type: string
+        required: true
+
+env:
+    VENV_BASE: "WHL-TEST"
+    WHL_TEST_RESULTS: "whl-test-results"
+
+jobs:
+
+    BUILD-TEST-WHL:
+
+        runs-on: ${{ inputs.build_label }}
+        timeout-minutes: ${{ fromJson(inputs.timeout) }}
+
+        steps:
+
+            - name: checkout
+              id: checkout
+              uses: actions/checkout@v4
+              with:
+                fetch-depth: 0
+                ref: ${{ inputs.gitref }}
+                submodules: recursive
+
+            - name: setenv
+              id: setenv
+              uses: ./.github/actions/nm-set-env/
+              with:
+                hf_token: ${{ secrets.NM_HF_TOKEN }}
+                Gi_per_thread: ${{ inputs.Gi_per_thread }}
+
+            - name: set python
+              id: set_python
+              uses: ./.github/actions/nm-set-python/
+              with:
+                python: ${{ inputs.python }}
+                venv: ${{ env.VENV_BASE }}
+
+            - name: create testmo run
+              id: create_testmo_run
+              uses: ./.github/actions/nm-testmo-run-create/
+              if: success() || failure()
+              with:
+                testmo_url: https://neuralmagic.testmo.net
+                testmo_token: ${{ secrets.TESTMO_TEST_TOKEN }}
+                source: 'build-whl'
+
+            - name: python lint
+              id: lint
+              uses: ./.github/actions/nm-lint-python/
+
+            - name: build
+              id: build
+              uses: ./.github/actions/nm-build-vllm/
+              with:
+                Gi_per_thread: ${{ inputs.Gi_per_thread }}
+                python: ${{ inputs.python }}
+                venv: ${{ env.VENV_BASE }}
+
+            - name: build whl
+              id: build_whl
+              uses: ./.github/actions/nm-build-vllm-whl/
+              with:
+                python: ${{ inputs.python }}
+
+            - name: upload whl
+              uses: actions/upload-artifact@v4
+              if: success() || failure()
+              with:
+                name: ${{ steps.build_whl.outputs.whl }}
+                path: dist/${{ steps.build_whl.outputs.whl }}
+                retention-days: 15
+
+            - name: complete testmo run
+              uses: ./.github/actions/testmo-run-complete/
+              if: success() || failure()
+              with:
+                testmo_url: https://neuralmagic.testmo.net
+                testmo_token: ${{ secrets.TESTMO_TEST_TOKEN }}
+                testmo_run_id: ${{ steps.create_testmo_run.outputs.id }}