feat: CW through the Yaesu keyer — fifth CW engine

The radio has a keyer and a command to feed it (KY), so an FTDX10 needs no
WinKeyer and no second cable, exactly as the Icom CI-V and Flex CWX engines
already do. The rig keys with its own timing, which is why the spacing is right
where a PC keying a line through USB latency drifts.

Text is filtered to what the keyer can actually send: an unsupported byte does
not produce an error on a Yaesu, it can abort the whole buffer, so the rest of a
macro would vanish silently. It is then fed in 24-character pieces, waiting for
room between them — the rig DROPS what does not fit, again with no error, so a
contest CQ would lose its tail.

STOP is the honest gap. Yaesu documents no buffer-clear, so it drops the
transmitter (TX0) instead: nothing queued reaches the air, which is what Escape
means to an operator. It deliberately does NOT send "KY0;" — a plausible-looking
clear that the rig would read as the CHARACTER zero and transmit.

A test caught a real one on the way: tabs and newlines were dropped as
"unsupported", gluing the words either side together, so a macro written on two
lines went out as CQCQ. Whitespace now becomes a word gap before filtering.

Settings warn when the Yaesu keyer is selected without the Yaesu CAT backend —
otherwise it simply never keys, with nothing on screen saying why.

Send path follows the CAT reference and how Hamlib drives these rigs; NOT yet
verified on the air.

internal/cat/yaesu_cw.go

@@ -0,0 +1,128 @@
+package cat
+
+// CW keying through the Yaesu's own keyer — the KY command.
+//
+// This is the fifth CW engine, alongside WinKeyer, the DTR/RTS line keyer, the
+// Icom CI-V keyer and FlexRadio's CWX. The point is the same in each: no extra
+// hardware, no second cable. The radio holds the text and keys it with its own
+// timing, which is why the character spacing is perfect where a PC keying a line
+// through USB latency is not.
+//
+//	KY;           → KY0;  buffer has room / KY1; buffer full
+//	KY <text>;            queue up to 24 characters
+//
+// The leading SPACE after KY is part of the command, not padding.
+//
+// Verified on: nothing yet — the send path follows the FTDX10/FTDX101 CAT
+// reference and the way Hamlib drives these rigs. STOP is the uncertain half:
+// Yaesu documents no way to clear the buffer, so see StopCW.
+
+import (
+	"fmt"
+	"strings"
+	"time"
+)
+
+// yaesuCWChunk is the most characters one KY command accepts. Longer text is
+// split and fed as the rig drains its buffer.
+const yaesuCWChunk = 24
+
+// yaesuCWAllowed is what the rig's keyer can actually send. Anything else is
+// dropped rather than passed through: an unsupported byte can abort the whole
+// buffer, losing the rest of the message with no error anywhere.
+const yaesuCWAllowed = "ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789 /?.,-=+:;()"
+
+// SendCW queues a message on the rig's keyer.
+//
+// Text is filtered, upper-cased and fed in 24-character pieces, waiting for room
+// between pieces. Without that wait a long macro would silently lose its tail:
+// the rig drops what does not fit rather than reporting an error.
+func (y *Yaesu) SendCW(text string) error {
+	msg := filterYaesuCW(text)
+	if msg == "" {
+		return nil
+	}
+	for len(msg) > 0 {
+		n := yaesuCWChunk
+		if len(msg) < n {
+			n = len(msg)
+		}
+		chunk := msg[:n]
+		msg = msg[n:]
+		if err := y.waitCWBuffer(4 * time.Second); err != nil {
+			return err
+		}
+		y.mu.Lock()
+		err := y.write("KY " + chunk + ";")
+		y.mu.Unlock()
+		if err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+// StopCW aborts the message being sent.
+//
+// Yaesu documents no buffer-clear, so this drops the transmitter instead: TX0
+// takes the rig out of transmit, which is what an operator pressing Escape
+// actually wants. Anything still queued is not keyed on the air.
+//
+// It deliberately does NOT send "KY0;" — a plausible-looking clear that the rig
+// would read as the CHARACTER zero and dutifully send. A wrong guess here
+// transmits, which is worse than an imperfect abort.
+func (y *Yaesu) StopCW() error {
+	y.mu.Lock()
+	defer y.mu.Unlock()
+	if y.port == nil {
+		return fmt.Errorf("yaesu: not connected")
+	}
+	return y.write("TX0;")
+}
+
+// waitCWBuffer blocks until the keyer reports room, or the deadline passes.
+// A rig that never answers the status query is not a reason to refuse to send —
+// we go ahead once, and the worst case is the truncation we were avoiding.
+func (y *Yaesu) waitCWBuffer(within time.Duration) error {
+	deadline := time.Now().Add(within)
+	for {
+		y.mu.Lock()
+		if y.port == nil {
+			y.mu.Unlock()
+			return fmt.Errorf("yaesu: not connected")
+		}
+		r, err := y.ask("KY;")
+		y.mu.Unlock()
+		if err != nil {
+			debugLog.Printf("yaesu cw: KY status query failed (%v) — sending anyway", err)
+			return nil
+		}
+		if len(r) >= 3 && r[2] == '0' {
+			return nil // room in the buffer
+		}
+		if time.Now().After(deadline) {
+			return fmt.Errorf("yaesu: the keyer buffer stayed full for %s", within)
+		}
+		time.Sleep(50 * time.Millisecond)
+	}
+}
+
+// filterYaesuCW upper-cases and strips what the keyer cannot send.
+func filterYaesuCW(text string) string {
+	var b strings.Builder
+	for _, r := range strings.ToUpper(text) {
+		// Any whitespace becomes a word gap FIRST. Dropping tabs and newlines as
+		// "not in the allowed set" glued the words either side together: a macro
+		// written on two lines went out as CQCQ.
+		if r == '\t' || r == '\n' || r == '\r' {
+			b.WriteByte(' ')
+			continue
+		}
+		if strings.ContainsRune(yaesuCWAllowed, r) {
+			b.WriteRune(r)
+		}
+	}
+	// Runs of spaces become one: the rig sends each as a word gap, so three in a
+	// row is three gaps and the message crawls.
+	return strings.Join(strings.Fields(b.String()), " ")
+}