feat: add empty-output retry and Vercel auto-continue support

- Auto-retry Chat/Responses streams once when upstream output is empty but not content-filtered, reusing session/token/PoW and appending a regeneration suffix to the prompt
- Wire DeepSeek continue API into Vercel streams for multi-round thinking output exhaustion
- Defer empty-output errors in stream finalizers to enable synthetic retry; only surface failure when the retry budget is exhausted
- Track content_filter stops to avoid retry on filtered outputs
- Add comprehensive tests for stream/non-stream retry, Responses retry, and content_filter no-retry
- Update prompt-compatibility.md documentation

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

internal/httpapi/openai/shared/empty_retry.go

@@ -0,0 +1,45 @@
+package shared
+
+import "strings"
+
+const EmptyOutputRetrySuffix = "Previous reply had no visible output. Please regenerate the visible final answer or tool call now."
+
+func EmptyOutputRetryEnabled() bool {
+	return true
+}
+
+func EmptyOutputRetryMaxAttempts() int {
+	return 1
+}
+
+func ClonePayloadWithEmptyOutputRetryPrompt(payload map[string]any) map[string]any {
+	clone := make(map[string]any, len(payload))
+	for k, v := range payload {
+		clone[k] = v
+	}
+	original, _ := payload["prompt"].(string)
+	clone["prompt"] = AppendEmptyOutputRetrySuffix(original)
+	return clone
+}
+
+func AppendEmptyOutputRetrySuffix(prompt string) string {
+	prompt = strings.TrimRight(prompt, "\r\n\t ")
+	if prompt == "" {
+		return EmptyOutputRetrySuffix
+	}
+	return prompt + "\n\n" + EmptyOutputRetrySuffix
+}
+
+func UsagePromptWithEmptyOutputRetry(originalPrompt string, retryAttempts int) string {
+	if retryAttempts <= 0 {
+		return originalPrompt
+	}
+	parts := make([]string, 0, retryAttempts+1)
+	parts = append(parts, originalPrompt)
+	next := originalPrompt
+	for i := 0; i < retryAttempts; i++ {
+		next = AppendEmptyOutputRetrySuffix(next)
+		parts = append(parts, next)
+	}
+	return strings.Join(parts, "\n")
+}