Kernel Tracing Fix (#439)

* Update lib/rocprofiler-sdk/hsa/queue.cpp

- switch using the kernel_pkt.kernel_dispatch.completion_signal instead of interrupt signal for getting the dispatch time

* Update tests/kernel-tracing/validate.py

- add verification of total runtime collected in test_timestamps
  - the sum of the runtime of all the kernels in reproducible-runtime should be ~1 sec +/- 10%

* Remove include/rocprofiler-sdk/rocprofiler_plugin.h

* Update CI workflow

- update actions/cache@v3 -> v4
- actions/cache/save@v3 -> v4
- thollander/actions-comment-pull-request@v2 -> v2.4.3

* Update pytest.ini

- change default options to one that is more verbose

* Update tests/kernel-tracing/CMakeLists.txt

- skip test_total_runtime when Address or Thread Sanitizer enabled
  - overhead skews the results

* Update tests/kernel-tracing/validate.py

- separate test_total_runtime test

[ROCm/rocprofiler-sdk commit: aaff4976d2]

projects/rocprofiler-sdk/tests/kernel-tracing/validate.py

@@ -74,6 +74,20 @@ def test_timestamps(input_data):
         assert api_end <= itr["end_timestamp"]
 
 
+def test_total_runtime(input_data):
+    sdk_data = input_data["rocprofiler-sdk-json-tool"]
+
+    runtime_data = []
+    for itr in sdk_data["buffer_records"]["kernel_dispatches"]:
+        elapsed = itr["end_timestamp"] - itr["start_timestamp"]
+        runtime_data.append(elapsed)  # in nanoseconds
+
+    expected_runtime = 1.0e3  # one second in milliseconds
+
+    assert (sum(runtime_data) * 1.0e-6) >= (0.9 * expected_runtime)
+    assert (sum(runtime_data) * 1.0e-6) <= (1.1 * expected_runtime)
+
+
 def test_internal_correlation_ids(input_data):
     data = input_data
     sdk_data = data["rocprofiler-sdk-json-tool"]