Finds exact text in caption tracks without modifying them. Matching uses the
same normalization, composite-media boundaries, and time coordinates as the
targeted caption edit endpoint. Fuzzy alternatives are returned separately
as suggestions and are never reported as exact matches. A resolved match
means the wording was located; a later write can still fail authorization,
version, or edit-boundary checks.

When more than 10 exact matches exist, use the one-based occurrence
parameter to retrieve a specific later match.

Authentication and request validation failures apply to the whole request.
Missing, inaccessible, or otherwise unreadable media are reported as
per-media statuses without exposing whether an inaccessible ID exists.

Requires api token with one of the following permissions

Read all folder and media data

Tokens with the "Act with a team member's permissions" permission
(all:delegate_to_contact_permissions scope) can also be used.