perf: disable Sentry subprocess instrumentation and tracing

StdlibIntegration monkeypatches subprocess.Popen, adding span/breadcrumb
overhead to every subprocess call. Profiling showed ~342k samples (~8.9%
of total runtime) in sentry_sdk.integrations.stdlib and sentry_sdk.utils.
Disable subprocess instrumentation and set traces/profiles sample rates
to 0 since codeflash is a CLI tool that doesn't need performance tracing.
Error capturing (capture_exception, capture_message, LoggingIntegration)
remains fully functional.

Author: KRRT7

codeflash/telemetry/sentry.py

@@ -2,6 +2,7 @@
 
 import sentry_sdk
 from sentry_sdk.integrations.logging import LoggingIntegration
+from sentry_sdk.integrations.stdlib import StdlibIntegration
 
 
 def init_sentry(*, enabled: bool = False, exclude_errors: bool = False) -> None:
@@ -15,13 +16,8 @@ def init_sentry(*, enabled: bool = False, exclude_errors: bool = False) -> None:
 
         sentry_sdk.init(
             dsn="https://4b9a1902f9361b48c04376df6483bc96@o4506833230561280.ingest.sentry.io/4506833262477312",
-            integrations=[sentry_logging],
-            # Set traces_sample_rate to 1.0 to capture 100%
-            # of transactions for performance monitoring.
-            traces_sample_rate=1.0,
-            # Set profiles_sample_rate to 1.0 to profile 100%
-            # of sampled transactions.
-            # We recommend adjusting this value in production.
-            profiles_sample_rate=1.0,
+            integrations=[sentry_logging, StdlibIntegration(subprocess_instrumentation=False)],
+            traces_sample_rate=0,
+            profiles_sample_rate=0,
             ignore_errors=[KeyboardInterrupt],
         )