[api] Inline ProtoSize::varint fast path for hot loop performance

Split ProtoSize::varint() into an always-inlined fast path (value < 128)
and a noinline slow path, eliminating indirect function calls on the BLE
advertisement encode hot path. Also force-inline calc_uint32() and
calc_length() so the compiler fully inlines size calculations per
advertisement instead of emitting out-of-line calls.

Verified via Xtensa disassembly: eliminates ~80 indirect function calls
per 16-advertisement batch flush (3 calc + varint per advertisement).
Total hot path code reduced from 24 to 21 functions.

esphome/components/api/proto.cpp

@@ -8,6 +8,19 @@ namespace esphome::api {
 
 static const char *const TAG = "api.proto";
 
+uint32_t ProtoSize::varint_slow_(uint32_t value) {
+  // value is guaranteed >= 128 here (fast path handled inline)
+  if (value < 16384) {
+    return 2;  // 14 bits
+  } else if (value < 2097152) {
+    return 3;  // 21 bits
+  } else if (value < 268435456) {
+    return 4;  // 28 bits
+  } else {
+    return 5;  // 32 bits (maximum for uint32_t)
+  }
+}
+
 #ifdef USE_API_VARINT64
 optional<ProtoVarInt> ProtoVarInt::parse_wide(const uint8_t *buffer, uint32_t len, uint32_t *consumed,
                                               uint32_t result32) {