feat(studio): default to 8 vCPU and 16 GB with one instance

[evanlowe]

veadk studio deploy now configures the Studio function with 8 vCPU, 16 GB memory, a minimum of 1 instance and a maximum of 1 instance on both Volcengine and BytePlus

The defaults apply to new deployments and same-name redeployments, including existing functions with smaller resources or a higher instance limit. Resource overrides remain optional in the shared deployment API, preserving ordinary Agent deployment defaults and unspecified existing settings

Validation:

• 175 related deployment, update and image-deployment unit tests passed after syncing with the latest main branch
• Pre-commit passed: Ruff checks, formatting and secret scanning
• Real cloud tests ran in parallel on Volcengine cn-beijing and BytePlus ap-southeast-1: initial and same-name source deployments completed successfully, including scheduler, worker and timers
• On each provider, the test function was temporarily changed to 2 vCPU / 4 GB and min 0 / max 2; redeployment preserved the application and function IDs and restored 8 vCPU / 16 GB and min 1 / max 1
• Home and configuration endpoints returned 200; real SSO login and authenticated access after redeployment passed on both providers
• Temporary applications, functions, timers and their SSO callback registrations were removed after verification

Cloud validation used veadk studio deploy --from-source. Large uploads hit the existing 300-second socket timeout, so the test processes temporarily used 1800 seconds; this PR does not change upload timeouts. The published PyPI distribution also had an existing startup incompatibility with the current Studio launcher, so successful runtime validation used the current source package