Kill off olm-loader

Now that Olm needs to be inited asynchronously anyway, we can just
pass the options to Olm.init(), and as long as we do that before we
start the js-sdk, we're all good.

This will means the olm js is now part of the main bundle but since
it's now just a wrapper around the wasm, this is probably faster.

Also add the directwatch flag to olm.wasm because otherwise it
doesn't seem to copy the file in watch mode...

src/vector/index.js

@@ -54,6 +54,8 @@ import SettingsStore, {SettingLevel} from "matrix-react-sdk/lib/settings/Setting
 import Tinter from 'matrix-react-sdk/lib/Tinter';
 import SdkConfig from "matrix-react-sdk/lib/SdkConfig";
 
+import Olm from 'olm';
+
 import rageshake from "matrix-react-sdk/lib/rageshake/rageshake";
 
 import CallHandler from 'matrix-react-sdk/lib/CallHandler';
@@ -227,6 +229,20 @@ async function loadApp() {
 
     window.addEventListener('hashchange', onHashChange);
 
+    /* Start loading Olm. Note that we *don't* wait for this to load: the
+     * js-sdk will also call this and actually wait on the promise before it
+     * tries to use the library. It can be loading its wasm while the rest of
+     * the app loads though.
+     *
+     * We also need to tell the Olm js to look for its wasm file at the same
+     * level as index.html. It really should be in the same place as the js,
+     * ie. in the bundle directory, to avoid caching issues, but as far as I
+     * can tell this is completely impossible with webpack.
+     */
+    Olm.init({
+        locateFile: () => 'olm.wasm',
+    });
+
     await loadLanguage();
 
     const fragparts = parseQsFromFragment(window.location);