Add --fast argument to enable experimental optimizations.

Optimizations that might break things/lower quality will be put behind
this flag first and might be enabled by default in the future.

Currently the only optimization is float8_e4m3fn matrix multiplication on
4000/ADA series Nvidia cards or later. If you have one of these cards you
will see a speed boost when using fp8_e4m3fn flux for example.

comfy/model_management.py

@@ -1048,6 +1048,16 @@ def should_use_bf16(device=None, model_params=0, prioritize_performance=True, ma
 
     return False
 
+def supports_fp8_compute(device=None):
+    props = torch.cuda.get_device_properties(device)
+    if props.major >= 9:
+        return True
+    if props.major < 8:
+        return False
+    if props.minor < 9:
+        return False
+    return True
+
 def soft_empty_cache(force=False):
     global cpu_state
     if cpu_state == CPUState.MPS: