fal video: route-declared limits, text_to_video at planning, and a generic adapter #94

• Route-declared video limits. fx-routes-v1 has no member for a route's aspect ratios, minimum duration, default tier or a tier's size, so fal-queue keeps them as constants (9:16 and 16:9, 3 seconds, 720p at 9:16, short sides 360 to 2160) and its check does not judge the size of another tier. A capped route (cap_usd alone) has neither tiers nor max_units, so a capped fal video route is held to the adapter's defaults (360p, 720p, 1080p, 4k; 3 to 10 seconds).
• text_to_video while planning. A step without a first_frame implies no feature, so on a route that does not declare text_to_video it is refused only when the call is sent (at $0, as before). gg-video does not judge it.
• A generic fal video adapter. fal's video request shapes diverge by endpoint (Grida Models v1.20261013.6, 12 endpoints): the first frame is image_url on 6 and start_image_url on 1; duration is an integer on 3, a string "4" to "30" (or auto) on 3, and "4s", "6s" or "8s" on 6; no endpoint records an end frame. fal-queue fits only endpoints with the built-in route's shape.

Found while building the AI provider contract (plan, #69, #79).