Fix the five library adoption review findings

internal/framework/processreport/llm_test.go

@@ -2,6 +2,7 @@ package processreport
 
 import (
 	"encoding/json"
+	"fmt"
 	"strings"
 	"testing"
 	"time"
@@ -54,3 +55,72 @@ func TestReportSummaryAvailability(t *testing.T) {
 		t.Fatalf("initialized empty summary: %+v", got)
 	}
 }
+
+func BenchmarkExecutionReports(b *testing.B) {
+	for _, n := range []int{1000, 10000, 20000} {
+		snapshot := llm.ObservationSnapshot{}
+		for i := 0; i < n; i++ {
+			snapshot.Calls = append(snapshot.Calls, llm.LogicalCall{CallID: uint64(i + 1), Prompt: structuredoutput.PromptMetadata{BackendID: "local"}, Context: structuredoutput.DiagnosticContext{ModuleInstance: "grammar", Role: "proposal"}})
+			snapshot.Events = append(snapshot.Events, llm.ExecutionObservation{CallID: uint64(i + 1), Kind: string(pk.ExecutionFinished), Attempts: 1})
+		}
+		output := &runner.RunOutput{Observations: snapshot}
+		b.Run(fmt.Sprintf("execution/%d", n), func(b *testing.B) {
+			for i := 0; i < b.N; i++ {
+				BuildLLMExecution(snapshot)
+			}
+		})
+		b.Run(fmt.Sprintf("utilization/%d", n), func(b *testing.B) {
+			for i := 0; i < b.N; i++ {
+				BuildUtilization(output, map[string]config.Backend{"local": {Concurrency: 1}})
+			}
+		})
+	}
+}
+
+func TestIndexedReportsPreserveOrderAndPartialUsage(t *testing.T) {
+	snapshot := llm.ObservationSnapshot{
+		Calls: []llm.LogicalCall{
+			{CallID: 3, Context: structuredoutput.DiagnosticContext{Role: "validation", ModuleInstance: "grammar_2"}, Prompt: structuredoutput.PromptMetadata{BackendID: "b"}},
+			{CallID: 1, Context: structuredoutput.DiagnosticContext{Role: "proposal", ModuleInstance: "grammar_1"}, Prompt: structuredoutput.PromptMetadata{BackendID: "a"}},
+			{CallID: 2, Context: structuredoutput.DiagnosticContext{Role: "validation", ModuleInstance: "grammar_1"}, Prompt: structuredoutput.PromptMetadata{BackendID: "a"}},
+		},
+		Events: []llm.ExecutionObservation{
+			{CallID: 1, Kind: string(pk.ExecutionFinished), Attempts: 1, Usage: llm.Usage{Total: llm.TokenCount{Value: 7, Known: true, Complete: true}}},
+			{CallID: 3, Kind: string(pk.ExecutionFinished), Attempts: 1},
+			{CallID: 2, Kind: string(pk.ExecutionFinished), Attempts: 1},
+		},
+	}
+	raw, err := json.Marshal(BuildLLMExecution(snapshot))
+	if err != nil {
+		t.Fatal(err)
+	}
+	var execution struct {
+		Calls  []executionCall  `json:"calls"`
+		Events []executionEvent `json:"events"`
+	}
+	if err := json.Unmarshal(raw, &execution); err != nil {
+		t.Fatal(err)
+	}
+	for i, call := range snapshot.Calls {
+		if execution.Calls[i].CallID != call.CallID {
+			t.Fatal("call order changed")
+		}
+	}
+	for i, event := range snapshot.Events {
+		if execution.Events[i].CallID != event.CallID {
+			t.Fatal("event order changed")
+		}
+	}
+	if execution.Calls[1].Summary.Usage.Total.Value != 7 || !execution.Calls[1].Summary.Usage.Total.Complete || execution.Calls[0].Summary.Usage.Total.Known {
+		t.Fatalf("call usage: %+v", execution.Calls)
+	}
+	output := &runner.RunOutput{Observations: snapshot}
+	utilization := BuildUtilization(output, map[string]config.Backend{"a": {}, "b": {}})
+	total := utilization.Backends["a"].Usage.Total
+	if total.Value != 7 || !total.Known || total.Complete || utilization.Backends["a"].ProviderAttempts != 2 {
+		t.Fatalf("partial backend usage: %+v", utilization)
+	}
+	if len(utilization.Modules) != 2 || utilization.Modules[1].ProposalLLMCalls != 1 || utilization.Modules[1].ValidationLLMCalls != 1 {
+		t.Fatalf("module grouping: %+v", utilization.Modules)
+	}
+}