fix: Resolve Docker build pip install failure

- Create requirements-docker.txt with minimal dependencies for IRIS compilation
- Remove invalid PyTorch 2.7.0 version requirement that doesn't exist
- Eliminate heavy ML dependencies that cause Docker build memory issues
- Keep only essential IRIS and database connectivity packages
- Update Dockerfile to use requirements-docker.txt instead of full requirements.txt
- This fixes the Docker build failure: 'pip3 install -r requirements.txt' step

The Docker CI now focuses on ObjectScript/ZPM compilation rather than full ML environment setup.

Author: tom-dyar

Dockerfile

@@ -27,7 +27,7 @@ ENV PATH "/usr/irissys/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sb
 COPY .iris_init /home/irisowner/.iris_init
 
 RUN --mount=type=bind,src=.,dst=. \
-    pip3 install -r requirements.txt && \
+    pip3 install -r requirements-docker.txt && \
     iris start IRIS && \
 	iris session IRIS < iris.script && \
     ([ $TESTS -eq 0 ] || iris session iris -U $NAMESPACE "##class(%ZPM.PackageManager).Shell(\"test $MODULE -v -only\",1,1)") && \

requirements-docker.txt

@@ -0,0 +1,17 @@
+# Docker-specific minimal requirements for IRIS ZPM compilation
+# This avoids heavy ML dependencies that cause build failures
+
+# Core IRIS dependencies only
+intersystems-irispython==5.1.2
+sqlalchemy>=2.0.0
+
+# Essential utilities
+requests>=2.31.0
+python-dotenv>=1.0.0
+PyYAML>=6.0
+
+# Database connectivity (minimal)
+jaydebeapi>=1.2.3
+
+# Testing (minimal)
+pytest>=7.4.0,<8.0.0