Adds a self-contained API workflow at
tests/inference/graphs/image_blend_channel_mismatch.json that exercises
the previously-broken RGB + RGBA blend case end-to-end:
EmptyImage (RGB, red) -> image1
EmptyImage (RGB, blue)
+ SolidMask (0.5) -> JoinImageWithAlpha (RGBA) -> image2
-> ImageBlend (3ch + 4ch)
-> SaveImage
The workflow uses no model checkpoints and runs in seconds on CPU. With
the fix it produces a 4-channel RGBA output (alpha preserved); the
previous behavior would have silently dropped the alpha channel via
node_helpers.image_alpha_fix.
Verified end-to-end against a local ComfyUI instance: workflow executes
successfully, output PNG is RGBA, center pixel is (127, 0, 127, 191) -
red and blue blended at 0.5 with alpha 0.5*1.5=0.75 (191/255).
* Add inference tests
* Clean up
* Rename test graph file
* Add readme for tests
* Separate server fixture
* test file name change
* Assert images are generated
* Clean up comments
* Add __init__.py so tests can run with command line `pytest`
* Fix command line args for pytest
* Loop all samplers/schedulers in test_inference.py
* Ci quality workflows compare (#1)
* Add image comparison tests
* Comparison tests do not pass with empty metadata
* Ensure tests are run in correct order
* Save image files with test name
* Update tests readme
* Reduce step counts in tests to ~halve runtime
* Ci quality workflows build (#2)
* Add build test github workflow
