Extends quantization predicate with config (#476)

Adds config parameter to quantization predicate
Enables fine-grained quantization control
Supports per-parameter quantization strategies
Improves flexibility in model quantization configuration

Author: robbiemu

mlx_lm/utils.py

@@ -496,7 +496,7 @@ def wrapped_predicate(path, module):
             return False
         bool_or_params = True
         if quant_predicate is not None:
-            bool_or_params = quant_predicate(path, module)
+            bool_or_params = quant_predicate(path, module, config)
         if isinstance(bool_or_params, dict):
             quantized_config["quantization"][path] = bool_or_params
         elif fine_grained_config and bool_or_params: