🪙 refactor: Collected Usage & Anthropic Prompt Caching (#11319)

* 🔧 refactor: Improve token calculation in AgentClient.recordCollectedUsage

- Updated the token calculation logic to sum output tokens directly from all entries, addressing issues with negative values in parallel execution scenarios.
- Added comments for clarity on the usage of input tokens and output tokens.
- Introduced a new test file for comprehensive testing of the recordCollectedUsage function, covering various execution scenarios including sequential and parallel processing, cache token handling, and model fallback logic.

* 🔧 refactor: Anthropic `promptCache` handling in LLM configuration

* 🔧 test: Add comprehensive test for cache token handling in recordCollectedUsage

- Introduced a new test case to validate the handling of cache tokens across multiple tool calls in the recordCollectedUsage function.
- Ensured correct calculations for input and output tokens, including scenarios with cache creation and reading.
- Verified the expected interactions with token spending methods to enhance the robustness of the token management logic.

packages/api/src/endpoints/openai/config.anthropic.spec.ts

@@ -39,6 +39,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
               type: 'enabled',
               budget_tokens: 2000,
             },
+            promptCache: true,
           },
         },
         configOptions: {
@@ -87,6 +88,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
               type: 'enabled',
               budget_tokens: 3000,
             },
+            promptCache: true,
           },
         },
         configOptions: {
@@ -134,6 +136,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
               user_id: 'user123',
             },
             topK: 50,
+            promptCache: true,
           },
         },
         configOptions: {
@@ -175,6 +178,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
             metadata: {
               user_id: 'user456',
             },
+            promptCache: true,
           },
         },
         configOptions: {
@@ -187,7 +191,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
       });
     });
 
-    it('should apply custom headers without anthropic-beta for models that dont need it', () => {
+    it('should apply custom headers and promptCache for models that support caching', () => {
       const apiKey = 'sk-custom';
       const endpoint = 'Anthropic (via LiteLLM)';
       const options = {
@@ -218,6 +222,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
             metadata: {
               user_id: undefined,
             },
+            promptCache: true,
           },
         },
         configOptions: {
@@ -300,6 +305,9 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
           stream: true,
           topP: 0.9,
           maxTokens: 2048,
+          modelKwargs: {
+            promptCache: true,
+          },
           // temperature is dropped
           // modelKwargs.topK is dropped
           // modelKwargs.metadata is dropped completely
@@ -379,6 +387,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
             metadata: {
               user_id: 'searchUser',
             },
+            promptCache: true,
           },
         },
         configOptions: {
@@ -425,6 +434,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
               user_id: 'testUser',
             },
             topK: 40,
+            promptCache: true,
           },
         },
         configOptions: {
@@ -470,6 +480,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
             metadata: {
               user_id: 'addUser',
             },
+            promptCache: true,
             customParam1: 'value1', // Unknown params added to modelKwargs
             customParam2: 42,
           },
@@ -519,6 +530,7 @@ describe('getOpenAIConfig - Anthropic Compatibility', () => {
             metadata: {
               user_id: 'bothUser',
             },
+            promptCache: true,
             customParam: 'customValue',
             // topK is dropped
           },