Add a Dockerfile for tools using bugbug nlp stuff (#934)

* Add a Dockerfile for tools using bugbug nlp stuff

* Use the bugbug-base-nlp image for the similarity training task

Fixes #933

docker-compose.yml

@@ -6,6 +6,12 @@ services:
       dockerfile: infra/dockerfile.base
     image: mozilla/bugbug-base
 
+  bugbug-nlp:
+    build:
+      context: .
+      dockerfile: infra/dockerfile.base-nlp
+    image: mozilla/bugbug-base-nlp
+
   bugbug-commit-retrieval:
     build:
       context: .

infra/data-pipeline.yml

@@ -476,7 +476,7 @@ tasks:
         - bugs-retrieval
       payload:
         maxRunTime: 25200
-        image: mozilla/bugbug-base:${version}
+        image: mozilla/bugbug-base-nlp:${version}
         command:
           - bugbug-train-similarity
           - --algorithm

infra/dockerfile.base-nlp

@@ -0,0 +1,13 @@
+FROM mozilla/bugbug-base:latest
+
+# Setup dependencies in a cacheable step
+ADD extra-nlp-requirements.txt /
+
+RUN apt-get update && \
+		apt-get install -y --no-install-recommends gcc g++ libgomp1 && \
+		pip install --disable-pip-version-check --no-cache-dir -r /extra-nlp-requirements.txt && \
+		apt-get remove -y gcc g++ && \
+		apt-get autoremove -y && \
+		rm -rf /var/lib/apt/lists/*
+
+RUN python -m spacy download en_core_web_sm