Sets profiler backend to NVTX by default (#3238)

# Description

We found that performance overhead varies depending on the profiler
backend, even when profiling is not enabled.
For optimization, we changed the default profiler backend to NVTX.
Depending on the test case, we observed a performance improvement of
about 1–12%.

## Type of change

- Bug fix (non-breaking change which fixes an issue)

## Checklist

- [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with
`./isaaclab.sh --format`
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [ ] I have updated the changelog and the corresponding version in the
extension's `config/extension.toml` file
- [ ] I have added my name to the `CONTRIBUTORS.md` or my name already
exists there
- [x] I have done this task

apps/isaaclab.python.headless.kit

@@ -88,6 +88,9 @@ app.audio.enabled = false
 # Enable Vulkan - avoids torch+cu12 error on windows
 app.vulkan = true
 
+# Set profiler backend to NVTX by default
+app.profilerBackend = "nvtx"
+
 # Disables rate limit in runloop
 app.runLoops.main.rateLimitEnabled=false
 

apps/isaaclab.python.headless.rendering.kit

@@ -88,6 +88,9 @@ app.audio.enabled = false
 # Enable Vulkan - avoids torch+cu12 error on windows
 app.vulkan = true
 
+# Set profiler backend to NVTX by default
+app.profilerBackend = "nvtx"
+
 # Disables rate limit in runloop
 app.runLoops.main.rateLimitEnabled=false
 

apps/isaacsim_4_5/isaaclab.python.headless.kit

@@ -78,6 +78,9 @@ app.audio.enabled = false
 # Enable Vulkan - avoids torch+cu12 error on windows
 app.vulkan = true
 
+# Set profiler backend to NVTX by default
+app.profilerBackend = "nvtx"
+
 # hide NonToggleable Exts
 exts."omni.kit.window.extensions".hideNonToggleableExts = true
 exts."omni.kit.window.extensions".showFeatureOnly = false

apps/isaacsim_4_5/isaaclab.python.headless.rendering.kit

@@ -83,6 +83,9 @@ app.audio.enabled = false
 # Enable Vulkan - avoids torch+cu12 error on windows
 app.vulkan = true
 
+# Set profiler backend to NVTX by default
+app.profilerBackend = "nvtx"
+
 # disable replicator orchestrator for better runtime perf
 exts."omni.replicator.core".Orchestrator.enabled = false