refactor(llm): move queryParams off model.native to typed field

Promotes queryParams to a first-class ModelRef field used by Endpoint.baseURL,
so deployment-level URL query params (Azure api-version, OpenAI-compatible
provider knobs) live in a typed home instead of an opaque `native` bag.

Also removes write-only dead fields from `native`:

- openaiCompatibleProvider (set by family helper, never read)
- opencodeProviderID, opencodeModelID (set by opencode bridge + native session
  builder, never read)
- npm (set by opencode bridge, never read)

After this commit `model.native` only carries genuinely provider-specific
opaque options that no other adapter cares about (Bedrock's aws_credentials
+ aws_region for SigV4). Drops the now-dead ProviderShared.queryParams
helper. Updates AGENTS.md doc on native is implicit through the new schema
JSDoc.

packages/opencode/src/provider/llm-bridge.ts

@@ -134,14 +134,9 @@ export const toModelRef = (input: Input): ModelRef | undefined => {
     baseURL: baseURL(input, resolution, options),
     apiKey: apiKey(input, resolution, options),
     headers: headers(input, options),
+    queryParams: resolution.queryParams,
     capabilities: capabilities(input, resolution),
     limits: LLM.limits({ context: input.model.limit.context, output: input.model.limit.output }),
-    native: {
-      opencodeProviderID: input.provider.id,
-      opencodeModelID: input.model.id,
-      npm: input.model.api.npm,
-      ...(resolution.queryParams ? { queryParams: resolution.queryParams } : {}),
-    },
   })
 }
 

packages/opencode/src/session/llm-native.ts

@@ -249,11 +249,7 @@ export const request = Effect.fn("LLMNative.request")(function* (input: RequestI
     toolChoice: input.toolChoice,
     generation: input.generation,
     metadata: input.metadata,
-    native: {
-      opencodeProviderID: input.provider.id,
-      opencodeModelID: input.model.id,
-      ...input.native,
-    },
+    native: input.native,
   })
 })