optimize numpy dot product

In numpy 1.26

`a.dot(b.T)`
is 6.5 times slower than
`c = b.T.copy()`
`a.dot(c)`

Author: lhqing

ALLCools/integration/cca.py

@@ -72,7 +72,8 @@ def cca(
 
     # CCA decomposition
     model = TruncatedSVD(n_components=n_components, algorithm=svd_algorithm, random_state=random_state)
-    U = model.fit_transform(tf_data1.dot(tf_data2.T))
+    tf_data2_t = tf_data2.T.copy()
+    U = model.fit_transform(tf_data1.dot(tf_data2_t))
 
     # select dimensions with non-zero singular values
     sel_dim = model.singular_values_ != 0