vercel: Use proper model identifiers and add image support (#33377)

Follow up to previous PRs:
- Return `true` in `supports_images` - v0 supports images already
- Rename model id to match the exact version of the model `v0-1.5-md`
(For now we do not expose `sm`/`lg` variants since they seem not to be
available via the API)
- Provide autocompletion in settings for using `vercel` as a `provider`

Release Notes:

- N/A

crates/language_models/src/provider/vercel.rs

@@ -303,14 +303,14 @@ impl LanguageModel for VercelLanguageModel {
     }
 
     fn supports_images(&self) -> bool {
-        false
+        true
     }
 
     fn supports_tool_choice(&self, choice: LanguageModelToolChoice) -> bool {
         match choice {
-            LanguageModelToolChoice::Auto => true,
-            LanguageModelToolChoice::Any => true,
-            LanguageModelToolChoice::None => true,
+            LanguageModelToolChoice::Auto
+            | LanguageModelToolChoice::Any
+            | LanguageModelToolChoice::None => true,
         }
     }
 
@@ -398,7 +398,7 @@ pub fn count_vercel_tokens(
             }
             // Map Vercel models to appropriate OpenAI models for token counting
             // since Vercel uses OpenAI-compatible API
-            Model::VZero => {
+            Model::VZeroOnePointFiveMedium => {
                 // Vercel v0 is similar to GPT-4o, so use gpt-4o for token counting
                 tiktoken_rs::num_tokens_from_messages("gpt-4o", &messages)
             }