feat(relay): hosted-tool conversion fidelity, reasoning normalization, and billing usage integrity (#7137)

* feat(relaykit): preserve hosted tools across conversions

- add protocol-neutral hosted-tool DTOs, conversion metadata, and loss policies
- bridge citations, grounding metadata, and hosted-tool stream lifecycles
- document the public conversion behavior and channel policy controls

* refactor(relaykit): normalize reasoning and thinking intent

- centralize provider-neutral reasoning intent, effort, and budget mappings
- parse model suffixes at the host entry boundary while preserving provider-owned tails
- keep adaptive Claude thinking and explicit zero-token compatibility consistent

* fix(billing): preserve authoritative usage across relay hops

- carry native BillingUsage sidecars through direct and streamed protocol bridges
- merge partial and terminal usage monotonically with safe fallback settlement
- retain cache metadata, penultimate usage, and per-call Gemini tool surcharges

* feat(relay): bridge Responses with Claude and Gemini protocols

- add direct request, response, and stream converters across supported relay formats
- expose Claude count_tokens and Chat-to-Responses compatibility endpoints
- carry conversion diagnostics through the host while retaining the curated public goldens

* fix(relay): wire relaykit conversions into host channels

- connect handlers, adaptors, and channel settings to the standalone conversion layer
- keep model mapping, pricing identity, retries, and provider-specific suffix behavior aligned
- ignore local audit artifacts and retain focused public regression coverage

setting/model_setting/global.go

@@ -35,6 +35,7 @@ func (p ChatCompletionsToResponsesPolicy) IsChannelEnabled(channelID int, channe
 type GlobalSettings struct {
 	PassThroughRequestEnabled        bool                             `json:"pass_through_request_enabled"`
 	ThinkingModelBlacklist           []string                         `json:"thinking_model_blacklist"`
+	EffortTailModelIDs               []string                         `json:"effort_tail_model_ids"`
 	ChatCompletionsToResponsesPolicy ChatCompletionsToResponsesPolicy `json:"chat_completions_to_responses_policy"`
 }
 
@@ -45,6 +46,13 @@ var defaultOpenaiSettings = GlobalSettings{
 		"moonshotai/kimi-k2-thinking",
 		"kimi-k2-thinking",
 	},
+	EffortTailModelIDs: []string{
+		"gpt-5.1-codex-max",
+		"qwen-image-edit-max",
+		"qwen-max",
+		"stable-diffusion-3-medium",
+		"yi-medium",
+	},
 	ChatCompletionsToResponsesPolicy: ChatCompletionsToResponsesPolicy{
 		Enabled:     false,
 		AllChannels: true,
@@ -77,3 +85,26 @@ func ShouldPreserveThinkingSuffix(modelName string) bool {
 	}
 	return false
 }
+
+// ShouldPreserveEffortTail reports model IDs whose names already end in an
+// effort-like token and must not be treated as reasoning aliases.
+func ShouldPreserveEffortTail(modelName string) bool {
+	target := strings.TrimSpace(modelName)
+	if target == "" {
+		return false
+	}
+	bare := target
+	if slash := strings.LastIndex(bare, "/"); slash >= 0 {
+		bare = bare[slash+1:]
+	}
+	for _, entry := range globalSettings.EffortTailModelIDs {
+		entry = strings.TrimSpace(entry)
+		if entry == "" {
+			continue
+		}
+		if entry == target || entry == bare {
+			return true
+		}
+	}
+	return false
+}