bandaid for run-readme-pr-macos.yml incorrectly loading to MPS

as per #1416 torchchat on hosts without MPS (which is all github hosts which use kvm to virtualize MacOS, but not MPS) should choose CPU as "fast" device.  The logic is present (see discussion in #1416 ), but either not fully functional (that would be the easier one to fix, just print the result of get_device_str and fix the code!) or specifically ignored on load in torch/serialization.py (If this is the case, we're effectively looking at a core PyTorch issue....)

In the meantime, this bandaid just forces the use of CPU on MacOS tests, to make MacOS tests run on CPU -- labeit hsortcircuiting test/execution of the "fast" device logic.  Not ideal, but some testing beats no testing.

Author: mikekgfb

.github/workflows/run-readme-pr-macos.yml

@@ -33,7 +33,7 @@ jobs:
           sysctl machdep.cpu.core_count
           echo "::endgroup::"
 
-          .ci/scripts/run-docs readme
+          TORCHCHAT_DEVICE=cpu .ci/scripts/run-docs readme
   
           echo "::group::Completion"
           echo "tests complete"
@@ -68,7 +68,7 @@ jobs:
           sysctl machdep.cpu.core_count
           echo "::endgroup::"
 
-          .ci/scripts/run-docs quantization
+          TORCHCHAT_DEVICE=cpu .ci/scripts/run-docs quantization
   
           echo "::group::Completion"
           echo "tests complete"
@@ -103,7 +103,7 @@ jobs:
           sysctl machdep.cpu.core_count
           echo "::endgroup::"
 
-          .ci/scripts/run-docs gguf
+          TORCHCHAT_DEVICE=cpu .ci/scripts/run-docs gguf
 
           echo "::group::Completion"
           echo "tests complete"
@@ -137,7 +137,7 @@ jobs:
           sysctl machdep.cpu.core_count
           echo "::endgroup::"
 
-          .ci/scripts/run-docs advanced
+          TORCHCHAT_DEVICE=cpu .ci/scripts/run-docs advanced
 
           echo "::group::Completion"
           echo "tests complete"