Add browser pool core data source

[IlyaasK]

Summary

• register the read-only kernel_browser_pool data source
• support exact lookup by canonical ID or exact name
• preserve project scoping: data-source override, provider project_id, then API-key binding
• expose only the first durable fields: id, name, project_id, and size
• add generated Registry documentation and focused unit coverage

Why

Existing browser pools can already be imported into the resource when Terraform should manage them. This data source covers the separate read-only case: reference an existing pool without taking ownership of its lifecycle.

The implementation performs one durable SDK GET and verifies that the returned ID or name matches the exact selector. API fields are validated before entering Terraform state.

Intentionally unsupported

• browser/session runtime operations or runtime state
• browser-pool writes through the data source
• resolved profile and extension references
• remaining durable launch, warmup, viewport, and Chrome policy fields
• live acceptance coverage in this first stack slice

Those durable read fields and acceptance coverage land in the following small PRs. No existing resource or data-source behavior changes.

Tests run

• go test -short -timeout=2m ./...
• go vet ./...
• gofmt -l cmd internal
• bash scripts/check-docs.sh
• git diff --check origin/main...HEAD

Focused tests cover schema and registration, Framework state writes, ID/name lookup, explicit/default/unscoped project selection, unnamed pools, API errors, nil or malformed responses, and selector mismatches.

Acceptance status

Not run in this PR. Live ID/name lookup and project scoping are added to the opt-in acceptance suite at the end of this stack.

[cursor]

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

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 63f8898. Configure here.}