fix: propagate LLM timeout to AI Core client

Address review feedback:

MAJOR:
- AICoreClient now defaults to 5min timeout (matching Client)
- Add AICoreClient.WithTimeout() for explicit timeout control
- Client.WithAICore() inherits parent client's timeout
- Client.WithTimeout() propagates to aicore client if set

MINOR:
- Extract AICoreOpenAIAPIVersion constant for the hardcoded api-version
- Tighten IsAnthropicModel to only match 'anthropic--' prefix
  (SAP AI Core always uses this prefix for Anthropic models)

NIT:
- Use fmt.Sprintf for token generation in tests (robust for >9 calls)
- Add error checking in TestAICoreClient_TokenExpiry
- Add tests for WithTimeout propagation

llm/aicore.go

@@ -13,6 +13,10 @@ import (
 	"time"
 )
 
+// AICoreOpenAIAPIVersion is the API version used for OpenAI models through AI Core.
+// Update this when SAP AI Core releases a new stable version.
+const AICoreOpenAIAPIVersion = "2024-12-01-preview"
+
 // AICoreConfig holds SAP AI Core authentication and connection settings.
 type AICoreConfig struct {
 	ClientID      string
@@ -40,14 +44,22 @@ type deployment struct {
 }
 
 // NewAICoreClient creates a new AI Core client with the given configuration.
+// The client uses a default 5-minute timeout; use WithTimeout to customize.
 func NewAICoreClient(cfg AICoreConfig) *AICoreClient {
 	return &AICoreClient{
 		config:      cfg,
-		http:        &http.Client{Timeout: 30 * time.Second},
+		http:        &http.Client{Timeout: 5 * time.Minute},
 		deployments: make(map[string]deployment),
 	}
 }
 
+// WithTimeout sets the HTTP request timeout for AI Core calls.
+// This should be called during construction, before concurrent use.
+func (c *AICoreClient) WithTimeout(d time.Duration) *AICoreClient {
+	c.http.Timeout = d
+	return c
+}
+
 // getToken returns a valid OAuth token, refreshing if necessary.
 func (c *AICoreClient) getToken(ctx context.Context) (string, error) {
 	c.mu.RLock()
@@ -154,7 +166,6 @@ func (c *AICoreClient) getDeploymentURL(ctx context.Context, model string) (stri
 }
 
 func (c *AICoreClient) fetchDeployments(ctx context.Context, token string) error {
-
 	deployURL := strings.TrimRight(c.config.APIURL, "/") + "/v2/lm/deployments"
 	req, err := http.NewRequestWithContext(ctx, http.MethodGet, deployURL, nil)
 	if err != nil {
@@ -180,10 +191,10 @@ func (c *AICoreClient) fetchDeployments(ctx context.Context, token string) error
 
 	var deployResp struct {
 		Resources []struct {
-			ID               string `json:"id"`
-			DeploymentURL    string `json:"deploymentUrl"`
-			Status           string `json:"status"`
-			Details          struct {
+			ID            string `json:"id"`
+			DeploymentURL string `json:"deploymentUrl"`
+			Status        string `json:"status"`
+			Details       struct {
 				Resources struct {
 					BackendDetails struct {
 						Model struct {
@@ -326,8 +337,8 @@ func (c *AICoreClient) CompleteOpenAI(ctx context.Context, model string, message
 		return "", fmt.Errorf("marshal request: %w", err)
 	}
 
-	// AI Core uses /chat/completions?api-version=2024-12-01-preview for OpenAI models
-	chatURL := strings.TrimRight(deployURL, "/") + "/chat/completions?api-version=2024-12-01-preview"
+	// AI Core uses /chat/completions?api-version=<version> for OpenAI models
+	chatURL := strings.TrimRight(deployURL, "/") + "/chat/completions?api-version=" + AICoreOpenAIAPIVersion
 	req, err := http.NewRequestWithContext(ctx, http.MethodPost, chatURL, bytes.NewReader(data))
 	if err != nil {
 		return "", fmt.Errorf("create request: %w", err)
@@ -363,6 +374,7 @@ func (c *AICoreClient) CompleteOpenAI(ctx context.Context, model string, message
 }
 
 // IsAnthropicModel returns true if the model name indicates an Anthropic model.
+// SAP AI Core uses "anthropic--" prefix for Anthropic models (e.g., "anthropic--claude-3-5-sonnet").
 func IsAnthropicModel(model string) bool {
-	return strings.HasPrefix(model, "anthropic--") || strings.Contains(model, "claude")
+	return strings.HasPrefix(model, "anthropic--")
 }