fix(core): Filter out unactionable CEFSharp promise rejection error by default

[Lms24]

We have internal user reports about an uncaught promise rejection being thrown by a .NET library called CEFSharp. This lib can embed chromium in a .NET app, which I guess means, a web page can be displayed within a .NET app. Apparently there's some problem that spams this error to some of our users.

This error seems to happen quite frequently and it's not the first time we're hearing about it:

• "Non-Error promise rejection captured with value:"  #3440 (comment)
• UnhandledRejection: Non-Error promise rejection captured with value: Object Not Found Matching Id:3 #3796
• Stackoverflow: https://stackoverflow.com/questions/75536442/troubleshooting-non-error-promise-rejection-captured-with-value-object-not-fou
• Issue in DDog browser SDK repo: Large number of Uncaught "Object Not Found Matching" errors DataDog/browser-sdk#2715 (comment)
• A blog post suggesting this is related to a crawler: https://trackjs.com/javascript-errors/object-not-found-matching-id-methodname-paramcount/
• More context suggesting that the error mainly originates from an outlook crawler which uses CEFSharp: Observed `Object Not Found Matching Id` error in Sentry cefsharp/CefSharp#3633 (comment)

This was also already reported in the CEFSharp repo

IMHO we can just filter this error by default, so this PR adds the error to be ignored by our default inboundFiltersIntegration (which I still think has a completely misleading name by the way 😅)

[github-actions]

size-limit report 📦

Path	Size	% Change	Change
@sentry/browser	23.27 KB	+0.32%	+75 B 🔺
@sentry/browser - with treeshaking flags	21.94 KB	+0.33%	+73 B 🔺
@sentry/browser (incl. Tracing)	35.75 KB	+0.2%	+73 B 🔺
@sentry/browser (incl. Tracing, Replay)	72.98 KB	+0.1%	+73 B 🔺
@sentry/browser (incl. Tracing, Replay) - with treeshaking flags	63.38 KB	+0.13%	+78 B 🔺
@sentry/browser (incl. Tracing, Replay with Canvas)	77.29 KB	+0.1%	+76 B 🔺
@sentry/browser (incl. Tracing, Replay, Feedback)	89.78 KB	+0.08%	+69 B 🔺
@sentry/browser (incl. Feedback)	40.02 KB	+0.18%	+73 B 🔺
@sentry/browser (incl. sendFeedback)	27.88 KB	+0.26%	+73 B 🔺
@sentry/browser (incl. FeedbackAsync)	32.67 KB	+0.22%	+73 B 🔺
@sentry/react	25.94 KB	+0.28%	+72 B 🔺
@sentry/react (incl. Tracing)	38.58 KB	+0.2%	+76 B 🔺
@sentry/vue	27.47 KB	+0.26%	+72 B 🔺
@sentry/vue (incl. Tracing)	37.59 KB	+0.2%	+74 B 🔺
@sentry/svelte	23.43 KB	+0.33%	+77 B 🔺
CDN Bundle	24.41 KB	+0.24%	+58 B 🔺
CDN Bundle (incl. Tracing)	37.4 KB	+0.19%	+69 B 🔺
CDN Bundle (incl. Tracing, Replay)	72.6 KB	+0.1%	+68 B 🔺
CDN Bundle (incl. Tracing, Replay, Feedback)	77.96 KB	+0.09%	+64 B 🔺
CDN Bundle - uncompressed	71.69 KB	+0.18%	+128 B 🔺
CDN Bundle (incl. Tracing) - uncompressed	110.97 KB	+0.12%	+128 B 🔺
CDN Bundle (incl. Tracing, Replay) - uncompressed	225.04 KB	+0.06%	+128 B 🔺
CDN Bundle (incl. Tracing, Replay, Feedback) - uncompressed	238.26 KB	+0.06%	+128 B 🔺
@sentry/nextjs (client)	38.86 KB	+0.2%	+76 B 🔺
@sentry/sveltekit (client)	36.26 KB	+0.22%	+80 B 🔺
@sentry/node	163 KB	+0.05%	+73 B 🔺
@sentry/node - without tracing	99.88 KB	+0.08%	+73 B 🔺
@sentry/aws-serverless	127.38 KB	+0.06%	+77 B 🔺

View base workflow run

[mydea]

hmm, no strong feelings, but how important/common is this? Just thinking out loud, if it only happens infrequently, we could also just add it to the default inbound filters in relay? 🤔 it just adds some small bundle size, I am not against it but we should weigh this properly!

[Lms24]

Yeah I don't like the bundle size increase either. However, after some more googling, it looks like this happens quite frequently. Given that if we don't filter this out client-side, we still add the error to replays, I think we rather should eat the bytes 😬 I updated the PR description with various issues and posts I found.