Lower dynamic sampling rate target to avoid flakiness

This test for the "Skipped Samples" placeholder failed in CI:
<https://app.circleci.com/pipelines/github/DataDog/dd-trace-rb/15605/workflows/5b32e4bb-155e-4a94-bf8b-980599eaafae/jobs/568892/tests#failed-test-0>

I strongly suspect the issue is that the profiler was able to "keep up"
with the Ruby VM and thus never needed to skip samples.

In fact, I can reproduce the same behavior by raising the
`dynamic_sampling_rate_overhead_target_percentage` on my machine.

Thus, to reduce flakiness, I'm lowering the target for the dynamic
sampling rate from 2% to 0.1% which should be tight enough to
consistently cause the profiler to need to skip samples in this test.

spec/datadog/profiling/collectors/cpu_and_wall_time_worker_spec.rb

@@ -547,6 +547,7 @@
         # But the total amount of allocations recorded should match the number we observed, and thus we record the
         # remainder above the clamped value as a separate "Skipped Samples" step.
         context 'with a high allocation rate' do
+          let(:options) { { **super(), dynamic_sampling_rate_overhead_target_percentage: 0.1 } }
           let(:thread_that_allocates_as_fast_as_possible) { Thread.new { loop { BasicObject.new } } }
 
           after do