feat: capture Gemini cached content tokens in cost tracking

- Add CachedContentTokenCount to UsageMetadata parsing for both
  streaming (helpers.go) and non-streaming (gemini.go) requests
- CacheReadTokens now populated from Gemini cachedContentTokenCount
- Add uint32Ptr helper for nil-safe uint32 pointer creation

.pi-lens/turn-state.json

@@ -2,5 +2,5 @@
   "files": {},
   "turnCycles": 0,
   "maxCycles": 3,
-  "lastUpdated": "2026-04-27T01:09:48.183Z"
+  "lastUpdated": "2026-04-27T01:12:44.352Z"
 }

internal/providers/gemini.go

@@ -2,14 +2,14 @@ package providers
 
 import (
 	"context"
-	"time"
 	"encoding/json"
 	"fmt"
 	"strings"
+	"time"
 
+	"github.com/go-resty/resty/v2"
 	"gophergate/internal/config"
 	"gophergate/internal/models"
-	"github.com/go-resty/resty/v2"
 )
 
 type GeminiProvider struct {
@@ -54,10 +54,10 @@ type GeminiContent struct {
 }
 
 type GeminiPart struct {
-	Text             string                   `json:"text,omitempty"`
-	InlineData       *GeminiInlineData        `json:"inlineData,omitempty"`
-	FunctionCall     *GeminiFunctionCall      `json:"functionCall,omitempty"`
-	FunctionResponse *GeminiFunctionResponse  `json:"functionResponse,omitempty"`
+	Text             string                  `json:"text,omitempty"`
+	InlineData       *GeminiInlineData       `json:"inlineData,omitempty"`
+	FunctionCall     *GeminiFunctionCall     `json:"functionCall,omitempty"`
+	FunctionResponse *GeminiFunctionResponse `json:"functionResponse,omitempty"`
 }
 
 type GeminiInlineData struct {
@@ -265,9 +265,10 @@ func (p *GeminiProvider) ChatCompletion(ctx context.Context, req *models.Unified
 			FinishReason string `json:"finishReason"`
 		} `json:"candidates"`
 		UsageMetadata struct {
-			PromptTokenCount     uint32 `json:"promptTokenCount"`
-			CandidatesTokenCount uint32 `json:"candidatesTokenCount"`
-			TotalTokenCount      uint32 `json:"totalTokenCount"`
+			PromptTokenCount        uint32 `json:"promptTokenCount"`
+			CandidatesTokenCount    uint32 `json:"candidatesTokenCount"`
+			TotalTokenCount         uint32 `json:"totalTokenCount"`
+			CachedContentTokenCount uint32 `json:"cachedContentTokenCount"`
 		} `json:"usageMetadata"`
 	}
 
@@ -324,6 +325,7 @@ func (p *GeminiProvider) ChatCompletion(ctx context.Context, req *models.Unified
 			PromptTokens:     geminiResp.UsageMetadata.PromptTokenCount,
 			CompletionTokens: geminiResp.UsageMetadata.CandidatesTokenCount,
 			TotalTokens:      geminiResp.UsageMetadata.TotalTokenCount,
+			CacheReadTokens:  uint32Ptr(geminiResp.UsageMetadata.CachedContentTokenCount),
 		},
 	}
 
@@ -494,3 +496,10 @@ func (p *GeminiProvider) ChatCompletionStream(ctx context.Context, req *models.U
 
 	return ch, nil
 }
+
+func uint32Ptr(v uint32) *uint32 {
+	if v > 0 {
+		return &v
+	}
+	return nil
+}

internal/providers/helpers.go

@@ -134,10 +134,10 @@ func BuildOpenAIBody(request *models.UnifiedRequest, messagesJSON []interface{},
 }
 
 type openAIUsage struct {
-	PromptTokens            uint32 `json:"prompt_tokens"`
-	CompletionTokens        uint32 `json:"completion_tokens"`
-	TotalTokens             uint32 `json:"total_tokens"`
-	PromptTokensDetails     *struct {
+	PromptTokens        uint32 `json:"prompt_tokens"`
+	CompletionTokens    uint32 `json:"completion_tokens"`
+	TotalTokens         uint32 `json:"total_tokens"`
+	PromptTokensDetails *struct {
 		CachedTokens uint32 `json:"cached_tokens"`
 	} `json:"prompt_tokens_details"`
 	CompletionTokensDetails *struct {
@@ -308,6 +308,7 @@ func StreamGemini(ctx io.ReadCloser, ch chan<- *models.ChatCompletionStreamRespo
 						PromptTokens:     geminiChunk.UsageMetadata.PromptTokenCount,
 						CompletionTokens: geminiChunk.UsageMetadata.CandidatesTokenCount,
 						TotalTokens:      geminiChunk.UsageMetadata.TotalTokenCount,
+						CacheReadTokens:  uint32Ptr(geminiChunk.UsageMetadata.CachedContentTokenCount),
 					},
 				}
 			}