ci: install only the extras the tests use, and drop the stale asyncio_mode option

The unit-test workflow installed --all-extras, which after the packaging fixes
on this branch meant 316 packages: gcp, aws, azure, kfp, sagemaker, mlflow,
wandb and the genai stack on top of torch, tensorflow and a set of CUDA wheels.

An AST audit of every import in tests/ finds exactly four optional modules -
torch, tensorflow, keras and sklearn - and nothing else. tensorflow and
scikit-learn already come from the dev group and keras ships inside tensorflow,
so `pytorch` is the only extra the suite needs. The genai tests pass without
langchain, langgraph or openai installed (they mock the clients), and no test
imports mlflow, wandb, azure, kfp, sagemaker or google.cloud at all.

The three test jobs now install 188 packages instead of 316 - and instead of
the 289 they installed before this branch, so CI is lighter than it was rather
than heavier. No coverage is lost: the extras' own correctness is asserted by
tests/test_packaging_metadata.py, and the release workflow still performs a
full --all-extras install, which is where a broken extra actually matters.

Also removes `asyncio_mode = "auto"` from the pytest config. pytest-asyncio is
not a dependency, so the option was silently unknown and emitted a
PytestConfigWarning on every run. It was also a trap: a future bare
`async def test_` would have been collected and never awaited. The suite's four
async tests are driven by anyio's pytest plugin through `@pytest.mark.anyio`,
which works because anyio ships with starlette - verified by breaking an
assertion inside one and confirming it fails rather than passing vacuously.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01S1CErYvpggWAogJ9MDVqj4

Author: claude

.github/workflows/UTESTS.yml

@@ -64,7 +64,16 @@ jobs:
         run: |
           python -m pip install --upgrade pip
           pip install poetry
-          poetry install --no-interaction --with dev --all-extras
+          # Only the extras the tests actually exercise. An audit of every
+          # import in tests/ finds torch, tensorflow, keras and sklearn and
+          # nothing else; tensorflow and scikit-learn already come from the dev
+          # group, and keras ships with tensorflow, so `pytorch` is the only
+          # extra left. --all-extras additionally pulled gcp, aws, azure, kfp,
+          # sagemaker, mlflow, wandb and the genai stack - roughly 30 packages
+          # and a set of CUDA wheels that no test imports. The extras' own
+          # correctness is covered by tests/test_packaging_metadata.py and by
+          # the --all-extras install the release workflow still performs.
+          poetry install --no-interaction --with dev --extras pytorch
       - name: Run smoke tests
         run: |
           # Run a quick subset of tests for fast feedback
@@ -109,7 +118,16 @@ jobs:
         run: |
           python -m pip install --upgrade pip
           pip install poetry
-          poetry install --no-interaction --with dev --all-extras
+          # Only the extras the tests actually exercise. An audit of every
+          # import in tests/ finds torch, tensorflow, keras and sklearn and
+          # nothing else; tensorflow and scikit-learn already come from the dev
+          # group, and keras ships with tensorflow, so `pytorch` is the only
+          # extra left. --all-extras additionally pulled gcp, aws, azure, kfp,
+          # sagemaker, mlflow, wandb and the genai stack - roughly 30 packages
+          # and a set of CUDA wheels that no test imports. The extras' own
+          # correctness is covered by tests/test_packaging_metadata.py and by
+          # the --all-extras install the release workflow still performs.
+          poetry install --no-interaction --with dev --extras pytorch
 
       - name: Run tests - ${{ matrix.test-group }}
         run: |
@@ -182,7 +200,16 @@ jobs:
         run: |
           python -m pip install --upgrade pip
           pip install poetry
-          poetry install --no-interaction --with dev --all-extras
+          # Only the extras the tests actually exercise. An audit of every
+          # import in tests/ finds torch, tensorflow, keras and sklearn and
+          # nothing else; tensorflow and scikit-learn already come from the dev
+          # group, and keras ships with tensorflow, so `pytorch` is the only
+          # extra left. --all-extras additionally pulled gcp, aws, azure, kfp,
+          # sagemaker, mlflow, wandb and the genai stack - roughly 30 packages
+          # and a set of CUDA wheels that no test imports. The extras' own
+          # correctness is covered by tests/test_packaging_metadata.py and by
+          # the --all-extras install the release workflow still performs.
+          poetry install --no-interaction --with dev --extras pytorch
 
       - name: Run tests with coverage
         run: |

pyproject.toml

@@ -188,7 +188,10 @@ python_files = ["test_*.py"]
 python_functions = ["test_*"]
 # Sequential by default; CI/local can add `-n auto --dist=worksteal` for parallelism
 addopts = "-v --cov=flowyml --cov-report=term-missing"
-asyncio_mode = "auto"
+# No asyncio_mode here: pytest-asyncio is not a dependency, so the option was
+# silently unknown and only produced a PytestConfigWarning on every run. The
+# suite's async tests are driven by anyio's pytest plugin via
+# `@pytest.mark.anyio`, which works because anyio ships with starlette.
 # Markers for test organization
 markers = [
     "slow: marks tests as slow (deselect with '-m \"not slow\"')",