fix(vercelai): input token count

[obostjancic]

Makes sure that gen_ai.input_tokens value contains gen_ai.input_tokens.cached

Before submitting a pull request, please take a look at our
Contributing guidelines and verify:

• If you've added code that should be tested, please add tests.
• Ensure your code lints and the test suite passes (yarn lint) & (yarn test).
• Link an issue if there is one related to your pull request. If no issue is linked, one will be auto-generated and linked.

Closes #18577 (added automatically)

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	11,486	-	11,009	+4%
GET With Sentry	1,953	17%	1,817	+7%
GET With Sentry (error only)	7,672	67%	7,379	+4%
POST Baseline	1,154	-	1,106	+4%
POST With Sentry	587	51%	545	+8%
POST With Sentry (error only)	1,022	89%	990	+3%
MYSQL Baseline	4,008	-	3,810	+5%
MYSQL With Sentry	542	14%	422	+28%
MYSQL With Sentry (error only)	3,273	82%	3,107	+5%

View base workflow run