feat: batches with dry run, undo, and CSV import and export

- POST /_seo/api/v1/batch runs changes in one transaction: dry_run
  answers the diff and the conflicts (loops, chains, noindex on a
  published record, canonicals to redirected or noindex addresses) and
  rolls back; a refused change refuses the batch (D-12)
- undo of a change or a whole batch from History and through MCP, with
  the undo_stale, undo_loop and undo_gone refusals (D-13)
- CSV export and import of redirects, rules and records by URL: BOM,
  comma or semicolon, formula cells escaped, a dry-run preview, all or
  nothing with via import; the screen's apply token is HMAC-SHA256 over
  the file and a fingerprint of the data (transfer.stale)
- Import and export singleton with the seo_transfer widget action,
  History record actions undo and undo_batch, navigation entry
- MCP tools undo_change, undo_batch, apply_batch, export_csv, import_csv
- TestBatch, TestUndo, TestCSVTransfer, TestLoopRefused mcp and import

classes/csv.go

@@ -0,0 +1,145 @@
+package classes
+
+import (
+	"bytes"
+	"encoding/csv"
+	"errors"
+	"io"
+	"strings"
+)
+
+// BOM is the UTF-8 byte order mark export files start with, so Excel
+// opens Polish characters correctly; import strips it.
+const BOM = "\ufeff"
+
+// ErrUnreadable refuses a file that is not CSV with a header row
+// (transfer.unreadable).
+var ErrUnreadable = errors.New("golem15.seo: the file could not be read as CSV")
+
+// CSVDelimiter returns the delimiter rune of a setting: ";" or ",", comma
+// by default.
+func CSVDelimiter(d string) rune {
+	if strings.TrimSpace(d) == ";" {
+		return ';'
+	}
+	return ','
+}
+
+// formulaStart reports whether a cell would be read as a formula by a
+// spreadsheet: it starts with = + - @, a tab or a carriage return.
+func formulaStart(s string) bool {
+	if s == "" {
+		return false
+	}
+	switch s[0] {
+	case '=', '+', '-', '@', '\t', '\r':
+		return true
+	}
+	return false
+}
+
+// escapeCell prefixes a formula cell with an apostrophe (CSV formula
+// injection, T-32-24).
+func escapeCell(s string) string {
+	if formulaStart(s) {
+		return "'" + s
+	}
+	return s
+}
+
+// unescapeCell drops the apostrophe escapeCell added, so an exported file
+// imports unchanged.
+func unescapeCell(s string) string {
+	if strings.HasPrefix(s, "'") && formulaStart(s[1:]) {
+		return s[1:]
+	}
+	return s
+}
+
+// WriteCSV writes header and rows as UTF-8 with a BOM and the delimiter,
+// every formula cell escaped.
+func WriteCSV(header []string, rows [][]string, delimiter string) (string, error) {
+	var b bytes.Buffer
+	b.WriteString(BOM)
+	w := csv.NewWriter(&b)
+	w.Comma = CSVDelimiter(delimiter)
+	if err := w.Write(header); err != nil {
+		return "", err
+	}
+	for _, row := range rows {
+		cells := make([]string, len(row))
+		for i, c := range row {
+			cells[i] = escapeCell(c)
+		}
+		if err := w.Write(cells); err != nil {
+			return "", err
+		}
+	}
+	w.Flush()
+	return b.String(), w.Error()
+}
+
+// CSVRow is one data row of a file: its line number and its cells by
+// column name.
+type CSVRow struct {
+	Line   int
+	Fields map[string]string
+}
+
+// ReadCSV reads a file with a header row: the BOM is stripped, the
+// delimiter is the one given (detected from the header when empty), the
+// column names are trimmed and lowercased and the apostrophe of escaped
+// formula cells is dropped. Rows whose cells are all empty are skipped.
+// A file that is not CSV, has no header or was read with the wrong
+// delimiter answers ErrUnreadable.
+func ReadCSV(data, delimiter string) ([]string, []CSVRow, error) {
+	data = strings.TrimPrefix(data, BOM)
+	if strings.TrimSpace(data) == "" {
+		return nil, nil, ErrUnreadable
+	}
+	first, _, _ := strings.Cut(data, "\n")
+	comma := CSVDelimiter(delimiter)
+	if strings.TrimSpace(delimiter) == "" && strings.Count(first, ";") > strings.Count(first, ",") {
+		comma = ';'
+	}
+	r := csv.NewReader(strings.NewReader(data))
+	r.Comma = comma
+	r.FieldsPerRecord = -1
+	header, err := r.Read()
+	if err != nil {
+		return nil, nil, ErrUnreadable
+	}
+	for i, h := range header {
+		header[i] = strings.ToLower(strings.TrimSpace(h))
+	}
+	if len(header) == 1 && strings.ContainsAny(header[0], ",;") {
+		return nil, nil, ErrUnreadable
+	}
+	var rows []CSVRow
+	for {
+		rec, err := r.Read()
+		if errors.Is(err, io.EOF) {
+			break
+		}
+		if err != nil {
+			return nil, nil, ErrUnreadable
+		}
+		line, _ := r.FieldPos(0)
+		row := CSVRow{Line: line, Fields: map[string]string{}}
+		empty := true
+		for i, cell := range rec {
+			if i >= len(header) || header[i] == "" {
+				continue
+			}
+			cell = unescapeCell(strings.TrimSpace(cell))
+			if cell != "" {
+				empty = false
+			}
+			row.Fields[header[i]] = cell
+		}
+		if !empty {
+			rows = append(rows, row)
+		}
+	}
+	return header, rows, nil
+}