[QNNParam] Refactor the implmentation of QNNParam (#11011)

* The patch is to simplify the implmentation of QNNParam and make it more friendly to Python 2.x.

* Empty-Commit

* fix error about boolean value of Tensor with more than one value is ambiguous.

Author: billishyahao

python/tvm/relay/frontend/qnn_torch.py

@@ -28,17 +28,12 @@
 from .pytorch_utils import is_version_greater_than, getattr_attr_name
 
 
-class QNNParam:
+class QNNParam(object):
     """A placeholder for weight quantization parameters"""
 
     def __init__(self, weight, bias, scale, zero_point):
         self.weight = weight
-
-        if bias is not None:
-            self.bias = bias.detach().numpy()
-        else:
-            self.bias = None
-
+        self.bias = None if bias is None else bias.detach().numpy()
         self.scale = _expr.const(scale)
         self.zero_point = _expr.const(zero_point, dtype="int32")