feat: add language support and tests for English and Chinese seed data

- Introduced a new package `seedlang` to handle language parsing and constants for English and Chinese.
- Updated `Init` functions across various modules (kb, quickreply, skill, tag) to accept a language parameter.
- Implemented English seed data for FAQs, quick replies, skills, and tags, ensuring no Chinese text is present in English data.
- Added tests to verify that English seed data does not contain Chinese characters.
- Removed outdated README file from the kb test data directory.
- Created new test files for channel and kb to validate the absence of Chinese text in English seed data.

cmd/testdata/kb/kb_test.go

@@ -0,0 +1,34 @@
+package kb
+
+import (
+	"agent-desk/cmd/testdata/seedlang"
+	"regexp"
+	"testing"
+)
+
+var hanTextPattern = regexp.MustCompile(`\p{Han}`)
+
+func TestEnglishKnowledgeBaseTextDoesNotContainChineseText(t *testing.T) {
+	name, description, remark := faqKnowledgeBaseText(seedlang.English)
+	for _, value := range []string{name, description, remark} {
+		if hanTextPattern.MatchString(value) {
+			t.Fatalf("english knowledge base text contains Chinese text: %q", value)
+		}
+	}
+}
+
+func TestEnglishKnowledgeFAQSeedsDoNotContainChineseText(t *testing.T) {
+	seeds := knowledgeFAQSeeds(seedlang.English)
+	if len(seeds) == 0 {
+		t.Fatal("english FAQ seeds are empty")
+	}
+	for _, seed := range seeds {
+		values := []string{seed.Question, seed.Answer, seed.Remark}
+		values = append(values, seed.SimilarQuestions...)
+		for _, value := range values {
+			if hanTextPattern.MatchString(value) {
+				t.Fatalf("english FAQ seed contains Chinese text: %q", value)
+			}
+		}
+	}
+}