fix(android): Improve app start type detection with main thread timing

[markushi]

📜 Description

Improves app start type detection by utilizing the main thread idle callback.

Fixes #4920

The firstPost approach as utilized in the papa library does not seem to work on API level 35+ anymore (See square/papa#63)

The changes add:

• A new firstIdle field to track when the main thread becomes ready
• Enhanced logic in onActivityCreated() to detect warm starts when activity creation happens after the first main thread idle
• Comprehensive unit tests covering the new timing logic and edge cases
• Extended app start critical tests for determining warm/cold app start behavior (also see Add support for Android API level 36 in CI workflow #5030)

📝 Checklist

• I added GH Issue ID & Linear ID
• I added tests to verify the changes.
• No new PII added or SDK only sends newly added PII if sendDefaultPII is enabled.
• I updated the docs if needed.
• I updated the wizard if needed.
• Review from the native team if needed.
• No breaking change or entry added to the changelog.
• No breaking change for hybrid SDKs or communicated to hybrid SDKs.

🔮 Next steps

[github-actions]

Performance metrics 🚀

	Plain	With Sentry	Diff
Startup time	302.57 ms	320.71 ms	18.15 ms
Size	1.58 MiB	2.20 MiB	635.54 KiB

Baseline results on branch: main

Startup times

Revision	Plain	With Sentry	Diff
f634d01	359.58 ms	433.88 ms	74.30 ms
7314dbe	437.83 ms	505.64 ms	67.81 ms
fcec2f2	328.91 ms	387.75 ms	58.84 ms
ee747ae	382.73 ms	435.41 ms	52.68 ms
d15471f	343.13 ms	361.47 ms	18.34 ms
ee747ae	358.21 ms	389.41 ms	31.20 ms
bbc35bb	298.53 ms	372.17 ms	73.64 ms
fc5ccaf	322.49 ms	405.25 ms	82.76 ms
bbc35bb	324.88 ms	425.73 ms	100.85 ms
d15471f	379.40 ms	470.76 ms	91.36 ms

App size

Revision	Plain	With Sentry	Diff
f634d01	1.58 MiB	2.10 MiB	533.40 KiB
7314dbe	1.58 MiB	2.10 MiB	533.45 KiB
fcec2f2	1.58 MiB	2.12 MiB	551.50 KiB
ee747ae	1.58 MiB	2.10 MiB	530.95 KiB
d15471f	1.58 MiB	2.13 MiB	559.54 KiB
ee747ae	1.58 MiB	2.10 MiB	530.95 KiB
bbc35bb	1.58 MiB	2.12 MiB	553.01 KiB
fc5ccaf	1.58 MiB	2.13 MiB	557.54 KiB
bbc35bb	1.58 MiB	2.12 MiB	553.01 KiB
d15471f	1.58 MiB	2.13 MiB	559.54 KiB

Previous results on branch: fix/app-start-warm-detection

Startup times

Revision	Plain	With Sentry	Diff
12d79b2	324.79 ms	404.20 ms	79.41 ms
a0d0a05	303.68 ms	339.38 ms	35.70 ms

App size

Revision	Plain	With Sentry	Diff
12d79b2	1.58 MiB	2.20 MiB	635.40 KiB
a0d0a05	1.58 MiB	2.13 MiB	559.63 KiB

[markushi]

@sentry review

[markushi]

@sentry review

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

[cursor]

Incorrect boolean value in onActivityDestroyed sets wrong foreground state

Medium Severity

In onActivityDestroyed(), appLaunchedInForeground.setValue(true) is set when the app moves to background, but the original code set it to false. The comment explicitly states "if the app is moving into background", yet the value is inverted. This causes warm starts to potentially be misclassified as cold starts when a user returns from background very quickly (within the same millisecond as firstIdle), because the !appLaunchedInForeground.getValue() check that previously caught this case now fails, and the firstIdle timing check uses strict > comparison which can miss edge cases.