task(aws-lambda): Add python 3.15 support for aws-lambda-layer #8072

• add python 3.15 to the compatible runtimes of the aws-lambda-layer target in .craft.yml
• AND to --compatible-runtimes in scripts/aws/aws-deploy-local-layer.sh
• add py3.15 to aws_lambda tox env.

Docs should normally not be updated since runtimes are automatically fetched from AWS, and layer ARNs from the release-registry at build time; the fallback list may need to be updated since it currently stops at python3.14 (https://github.com/getsentry/sentry-docs/blob/a706b45968e0d9e685e5f940074866224a9d386b/src/build/awsLambdaRuntimes.ts#L4)

AWS Lambda integration also needs to be updated.

• sentry-python: add python3.15 to compatibleRuntimes of the aws-lambda-layer target in .craft.yml (9 of AWS's 15 maximum), and to . Check that PublishLayerVersion accepts python3.15 before merging: an invalid runtime fails the layer publish, and with it the release.
• sentry-python tests: the py3.15 classifier and tox envs already exist. Add py3.15 to the aws_lambda tox env once SAM has a python3.15 image (the tests use the host's Python version as the Lambda runtime), and verify the layer on x86_64 and arm64 like feat(aws-lambda): Enable Python 3.14 support to the Lambda layer #6916.
• sentry-docs: no change. Layer ARNs come from the release registry, and runtimes are fetched from AWS's lambda-runtimes.md at build time (src/build/awsLambdaRuntimes.ts). Only the fallback list used when that fetch fails is hardcoded, and it stops at python3.14.
• Sentry's AWS Lambda integration (getsentry/sentry): needs a change. SUPPORTED_RUNTIMES in src/sentry/integrations/aws_lambda/utils.py is hardcoded and filters the function list. It stops at python3.13 (and nodejs22.x), so python3.14 functions are already missing. Add python3.14 and python3.15, or match on the python3. / nodejs prefix so new runtimes work without a change.