Export fidelity: name padding, wOptions, as-caught catchRate (#206)

Auditing the reporter's PKHeX screenshots against the emitted bytes
turned up three fidelity bugs in the .sav export:

- Name fields on a templateless export stayed zero-filled after the
  $50 terminator; every real save the naming screen wrote $50-pads
  the tail, and the zero tail is what PKHeX rendered as "JOHN{}".
  encodeName now pads zero tail bytes with $50 (nonzero template
  bytes still survive untouched).
- wOptions was never written, dropping text speed / battle style /
  battle effects on export.  Templateless exports now pack it from
  save.options (the recomp's textSpeed 1/3/5 are pokered's exact
  values); with a template the cartridge's own byte still wins.
- The party/box catch-rate byte was always re-derived from the
  current species, but Gen1 freezes it at catch time (evolution does
  not update it), which is why PKHeX demanded "a preevolution catch
  rate".  Pokemon.new now stamps the as-caught catchRate and
  Evolution.apply's in-place mutation preserves it.

The reported emulator crash itself is not addressed here: the encoder
is byte-identical to the reporter's version, and a byte-level audit
(offsets vs Bulbapedia's save map, all three checksums, a real
gameplay save round-trip, pokered's LoadMapData regenerating every
zeroed cache) shows the current export is structurally valid.

src/save_convert/GenSave.lua

@@ -232,12 +232,17 @@ local function encodeName(buf, off, len, text)
     setByte(buf, off + i, charmap.byToken[ch] or charmap.byToken["?"] or 0x50)
     i, pos = i + 1, pos + clen
   end
-  -- Write exactly ONE $50 terminator and then STOP. The bytes after it are
-  -- left untouched: when encoding over a template they stay as the original
-  -- save's post-terminator padding (so an unchanged name round-trips
-  -- byte-identical), and on a templateless export they stay zero-filled. The
-  -- game reads a name only up to the first $50, so whatever follows is inert.
+  -- Write exactly ONE $50 terminator and then $50-pad the rest of the
+  -- field: every real save the naming screen ever wrote fills the tail
+  -- with $50, and a zero tail is what PKHeX renders as garbage glyphs
+  -- after the name ("JOHN{}", #206).  Template bytes that are NOT zero
+  -- stay untouched, so an unchanged name still round-trips
+  -- byte-identical and stale template data survives.
   if i < len then setByte(buf, off + i, 0x50) end
+  for j = i + 1, len - 1 do
+    local cur = buf[off + j + 1]
+    if cur == nil or cur:byte() == 0 then setByte(buf, off + j, 0x50) end
+  end
 end
 
 -- ------------------------------------------------------------------
@@ -656,6 +661,19 @@ function GenSave.encode(save, data, template)
   encodeName(buf, O.playerName, NAME_LENGTH, (save.player and save.player.name) or "RED")
   encodeName(buf, O.rivalName, NAME_LENGTH, (save.player and save.player.rival) or "BLUE")
   setU16be(buf, O.playerId, (save.player and save.player.id) or 0)
+  -- wOptions (engine/menus/main_menu.asm InitOptions): bit 7 = battle
+  -- effects OFF, bit 6 = SET style, bits 2-0 = text speed -- the recomp's
+  -- textSpeed 1/3/5 are pokered's exact FAST/MEDIUM/SLOW values
+  -- (SaveData.defaultOptions).  Templateless exports only: with a
+  -- template the byte survives untouched (the round-trip invariant), and
+  -- decode() never reads it back anyway.
+  if not src then
+    local opts = save.options or {}
+    local ob = (tonumber(opts.textSpeed) or 3) % 8
+    if opts.battleStyle == "set" then ob = bit.bor(ob, 0x40) end
+    if opts.animations == false then ob = bit.bor(ob, 0x80) end
+    setByte(buf, O.options, ob)
+  end
   setBcd(buf, O.money, 3, math.min(save.money or 0, 999999))
   setBcd(buf, O.coins, 2, math.min(save.coins or 0, 9999))