[PROTON] Correct misuse of strip (triton-lang#5716)

The `strip` method removes characters that match the input argument, but
it does not remove substrings that match the full input. For example,
previously, `"num_samples (inc)".strip("inc")` would result in
`um_samples` because the character `n` is part of the input argument.
Instead of using `strip`, the `replace` method can be used. However, in
this case, we can simply output the original metric name (i.e.,
"num_samples (inc)") since it’s more useful to know whether the metric
is inclusive or exclusive.

Author: Jokeren

third_party/proton/proton/viewer.py

@@ -249,7 +249,7 @@ def parse(metrics, filename, include=None, exclude=None, threshold=None, depth=1
         gf = filter_frames(gf, include, exclude, threshold, metrics[0])
         print(gf.tree(metric_column=metrics, expand_name=True, depth=depth, render_header=False))
         if print_sorted:
-            print("Sorted kernels by metric " + metrics[0].strip("(inc)"))
+            print("Sorted kernels by metric " + metrics[0])
             sorted_df = gf.dataframe.sort_values(by=[metrics[0]], ascending=False)
             for row in range(1, len(sorted_df)):
                 kernel_name = sorted_df.iloc[row]['name'][:100] + "..." if len(