global -> group (#1007)

* global -> group

* allow None for local_size in custom function

* lil local

* comment on shape

* fix cuda

* smart local cast

* better local heuristic

* fix ptx, and work_dim cleanup

* fix metal

* fix ops test

* fix openpilot jit

* no more optlocal

* might fix metal tests

* try metal now

* see generated metal code

* test free removal. REVERT THIS

* mergable

.github/workflows/test.yml

@@ -95,7 +95,7 @@ jobs:
       run: pip install -e '.[llvm,testing]' --extra-index-url https://download.pytorch.org/whl/cpu
     - name: Run Pytest
       run: ENABLE_METHOD_CACHE=1 LLVM=1 python -m pytest -s -v -n=auto test/
-  
+
   testclang:
     strategy:
       matrix:
@@ -207,11 +207,11 @@ jobs:
         python-version: 3.11
     - name: Install Dependencies
       run: pip install -e '.[metal,testing]'
-    - name: Run ops test
-      run: METAL=1 python -m pytest test/test_ops.py
-    # dtype test has issues on test_half_to_int8
     #- name: Run dtype test
-    #  run: METAL=1 python -m pytest test/test_dtype.py
+    #  run: DEBUG=4 METAL=1 python -m pytest test/test_dtype.py
+    - name: Run ops test
+      run: DEBUG=2 METAL=1 python -m pytest test/test_ops.py
+    # dtype test has issues on test_half_to_int8
 
   # disabled, this test is flaky
   testdocker: