*
     * @param request the chat request
     * @return the chat response
     * @throws LlmException if too many concurrent requests or interrupted
     */
    protected LlmChatResponse chatWithConcurrencyControl(final LlmChatRequest request) {
        if (concurrencyLimiter == null) {
            return chat(request);
        }
        if (logger.isDebugEnabled()) {

import org.codelibs.fess.entity.SearchRequestParams;
import org.codelibs.fess.helper.MarkdownRenderer;
import org.codelibs.fess.llm.ChatIntent;
import org.codelibs.fess.llm.IntentDetectionResult;
import org.codelibs.fess.llm.LlmChatResponse;
import org.codelibs.fess.llm.LlmClient;
import org.codelibs.fess.llm.LlmClientManager;
import org.codelibs.fess.llm.LlmException;
import org.codelibs.fess.llm.LlmMessage;
import org.codelibs.fess.llm.LlmStreamCallback;

            addIntentHistory(request, history);
        }

        // Implement abstract methods

        @Override
        public LlmChatResponse chat(final LlmChatRequest request) {
            // Capture the full request and user message for verification
            lastChatRequest = request;
            for (final LlmMessage msg : request.getMessages()) {