refactor: unify SDK model factories into internal/models

Move CreateModel, BuildReasoningOptions, ReasoningBudgetTokens and
related types from internal/agent to internal/models as NewSDKChatModel,
SDKModelConfig, etc. This eliminates duplicate ClientType constants and
centralises all Twilight AI SDK instance creation in a single package.

NewSDKEmbeddingModel now accepts a clientType parameter and dispatches
to the native Google embedding provider for google-generative-ai,
instead of always using the OpenAI-compatible endpoint.

internal/compaction/service.go

@@ -11,9 +11,9 @@ import (
 	"github.com/jackc/pgx/v5/pgtype"
 	sdk "github.com/memohai/twilight-ai/sdk"
 
-	"github.com/memohai/memoh/internal/agent"
 	"github.com/memohai/memoh/internal/db"
 	"github.com/memohai/memoh/internal/db/sqlc"
+	"github.com/memohai/memoh/internal/models"
 )
 
 // Service manages context compaction for bot conversations.
@@ -103,7 +103,7 @@ func (s *Service) doCompaction(ctx context.Context, logID pgtype.UUID, sessionUU
 
 	userPrompt := buildUserPrompt(priorSummaries, entries)
 
-	model := agent.CreateModel(agent.ModelConfig{
+	model := models.NewSDKChatModel(models.SDKModelConfig{
 		ClientType: cfg.ClientType,
 		BaseURL:    cfg.BaseURL,
 		APIKey:     cfg.APIKey,