refactor(ai-client): unify OpenAI-compatible path to AI SDK generateText

Eliminate the dual code path (raw fetch vs AI SDK) for text and vision.
All providers now go through createLanguageModel() + generateText(),
removing chatOpenAiCompatible/analyzeOpenAiCompatible, the manual Usage
type, summarizeUsage, and responseFormat plumbing from 8 call sites.

Key fix: @ai-sdk/openai v3 defaults to the Responses API (/responses);
DeepSeek only supports Chat Completions, so we use .chat() explicitly.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

lib/engine/director.ts

@@ -446,7 +446,7 @@ export async function directInsertBeat(
         content: buildInsertBeatUserMessage(session, freeformAction),
       },
     ],
-    { temperature: 0.9, responseFormat: "json_object", tag: "insert-beat" },
+    { temperature: 0.9, tag: "insert-beat" },
   );
 
   const parsed = parseJsonLoose<InsertBeatPartial>(raw);