fix: Report a rejected config section instead of aborting

makeApplication and Application::setup read config sections and throw
for one they reject. Neither call was wrapped, and main installs no
handler, so a missing [node_db] or an unparseable [node_seed] reached
the C++ runtime: "libc++abi: terminating due to uncaught exception"
followed by SIGABRT and exit 134. The [node_seed] case aborted after the
job queue had already started.

Both calls now sit in a try that writes the reason to stderr and exits
1.

run() is not reachable from a suite, so this is verified against a built
binary. A config with no [node_db] prints "xrpld: invalid configuration:
Missing [node_db] entry in configuration file" and exits 1. One whose
[node_seed] does not parse prints its own reason and exits 1. Both
aborted before.

src/xrpld/app/main/Main.cpp

@@ -804,11 +804,22 @@ run(int argc, char** argv)
         if (vm.contains("debug"))
             setDebugLogSink(logs->makeSink("Debug", beast::Severity::Trace));
 
-        auto app =
-            makeApplication(std::move(config), std::move(logs), std::make_unique<TimeKeeper>());
+        // Building the application and setting it up both read config
+        // sections, and throw for a section they reject.
+        std::unique_ptr<Application> app;
+        try
+        {
+            app =
+                makeApplication(std::move(config), std::move(logs), std::make_unique<TimeKeeper>());
 
-        if (!app->setup(vm))
-            return -1;
+            if (!app->setup(vm))
+                return -1;
+        }
+        catch (std::exception const& e)
+        {
+            std::cerr << "xrpld: invalid configuration: " << e.what() << std::endl;
+            return 1;
+        }
 
         // With our configuration parsed, ensure we have
         // enough file descriptors available: