feat(android-app): ChatGPT-style chat redesign and full LoRA adapter UI

Redesign chat screen to clean ChatGPT-like layout (no bubbles for assistant,
solid user bubbles) and add complete LoRA adapter management: picker sheet,
manager screen, model badges, real adapter registration from Void2377/Qwen.

Author: Siddhesh2377

examples/android/RunAnywhereAI/app/src/main/AndroidManifest.xml

@@ -26,6 +26,7 @@
         android:roundIcon="@mipmap/ic_launcher_round"
         android:supportsRtl="true"
         android:theme="@style/Theme.RunAnywhereAI"
+        android:windowSoftInputMode="adjustResize"
         android:largeHeap="true"
         android:usesCleartextTraffic="false"
         tools:targetApi="35">

examples/android/RunAnywhereAI/app/src/main/java/com/runanywhere/runanywhereai/MainActivity.kt

@@ -58,31 +58,26 @@ class MainActivity : ComponentActivity() {
         val initState by app.initializationState.collectAsState()
         val scope = rememberCoroutineScope()
 
-        Surface(
-            modifier = Modifier.fillMaxSize(),
-            color = MaterialTheme.colorScheme.background,
-        ) {
-            when (initState) {
-                is SDKInitializationState.Loading -> {
-                    InitializationLoadingView()
-                }
+        when (initState) {
+            is SDKInitializationState.Loading -> {
+                InitializationLoadingView()
+            }
 
-                is SDKInitializationState.Error -> {
-                    val error = (initState as SDKInitializationState.Error).error
-                    InitializationErrorView(
-                        error = error,
-                        onRetry = {
-                            scope.launch {
-                                app.retryInitialization()
-                            }
-                        },
-                    )
-                }
+            is SDKInitializationState.Error -> {
+                val error = (initState as SDKInitializationState.Error).error
+                InitializationErrorView(
+                    error = error,
+                    onRetry = {
+                        scope.launch {
+                            app.retryInitialization()
+                        }
+                    },
+                )
+            }
 
-                is SDKInitializationState.Ready -> {
-                    Log.i("MainActivity", "App is ready to use!")
-                    AppNavigation()
-                }
+            is SDKInitializationState.Ready -> {
+                Log.i("MainActivity", "App is ready to use!")
+                AppNavigation()
             }
         }
     }

examples/android/RunAnywhereAI/app/src/main/java/com/runanywhere/runanywhereai/data/ModelList.kt

@@ -6,9 +6,11 @@ import com.runanywhere.sdk.core.onnx.ONNX
 import com.runanywhere.sdk.core.types.InferenceFramework
 import com.runanywhere.sdk.llm.llamacpp.LlamaCPP
 import com.runanywhere.sdk.public.RunAnywhere
+import com.runanywhere.sdk.public.extensions.LoraAdapterCatalogEntry
 import com.runanywhere.sdk.public.extensions.ModelCompanionFile
 import com.runanywhere.sdk.public.extensions.Models.ModelCategory
 import com.runanywhere.sdk.public.extensions.Models.ModelFileDescriptor
+import com.runanywhere.sdk.public.extensions.registerLoraAdapter
 import com.runanywhere.sdk.public.extensions.registerModel
 import com.runanywhere.sdk.public.extensions.registerMultiFileModel
 
@@ -32,15 +34,15 @@ object ModelList {
         AppModel(id = "qwen2.5-0.5b-instruct-q6_k", name = "Qwen 2.5 0.5B Instruct Q6_K",
             url = "https://huggingface.co/Triangle104/Qwen2.5-0.5B-Instruct-Q6_K-GGUF/resolve/main/qwen2.5-0.5b-instruct-q6_k.gguf",
             framework = InferenceFramework.LLAMA_CPP, category = ModelCategory.LANGUAGE,
-            memoryRequirement = 600_000_000, supportsLoraAdapters = true),
+            memoryRequirement = 600_000_000),
         AppModel(id = "lfm2-350m-q4_k_m", name = "LiquidAI LFM2 350M Q4_K_M",
             url = "https://huggingface.co/LiquidAI/LFM2-350M-GGUF/resolve/main/LFM2-350M-Q4_K_M.gguf",
             framework = InferenceFramework.LLAMA_CPP, category = ModelCategory.LANGUAGE,
-            memoryRequirement = 250_000_000),
+            memoryRequirement = 250_000_000, supportsLoraAdapters = true),
         AppModel(id = "lfm2-350m-q8_0", name = "LiquidAI LFM2 350M Q8_0",
             url = "https://huggingface.co/LiquidAI/LFM2-350M-GGUF/resolve/main/LFM2-350M-Q8_0.gguf",
             framework = InferenceFramework.LLAMA_CPP, category = ModelCategory.LANGUAGE,
-            memoryRequirement = 400_000_000),
+            memoryRequirement = 400_000_000, supportsLoraAdapters = true),
         AppModel(id = "lfm2-1.2b-tool-q4_k_m", name = "LiquidAI LFM2 1.2B Tool Q4_K_M",
             url = "https://huggingface.co/LiquidAI/LFM2-1.2B-Tool-GGUF/resolve/main/LFM2-1.2B-Tool-Q4_K_M.gguf",
             framework = InferenceFramework.LLAMA_CPP, category = ModelCategory.LANGUAGE,
@@ -81,6 +83,60 @@ object ModelList {
             )),
     )
 
+    // LoRA Adapters (from Void2377/Qwen on HuggingFace — real standalone LoRA GGUF files)
+    private val loraAdapters = listOf(
+        LoraAdapterCatalogEntry(
+            id = "chat-assistant-lora",
+            name = "Chat Assistant",
+            description = "Enhances conversational chat ability",
+            downloadUrl = "https://huggingface.co/Void2377/Qwen/resolve/main/lora/chat_assistant-lora-Q8_0.gguf",
+            filename = "chat_assistant-lora-Q8_0.gguf",
+            compatibleModelIds = listOf("lfm2-350m-q4_k_m", "lfm2-350m-q8_0"),
+            fileSize = 690_176,
+            defaultScale = 1.0f,
+        ),
+        LoraAdapterCatalogEntry(
+            id = "summarizer-lora",
+            name = "Summarizer",
+            description = "Specialized for text summarization tasks",
+            downloadUrl = "https://huggingface.co/Void2377/Qwen/resolve/main/lora/summarizer-lora-Q8_0.gguf",
+            filename = "summarizer-lora-Q8_0.gguf",
+            compatibleModelIds = listOf("lfm2-350m-q4_k_m", "lfm2-350m-q8_0"),
+            fileSize = 690_176,
+            defaultScale = 1.0f,
+        ),
+        LoraAdapterCatalogEntry(
+            id = "translator-lora",
+            name = "Translator",
+            description = "Improves translation between languages",
+            downloadUrl = "https://huggingface.co/Void2377/Qwen/resolve/main/lora/translator-lora-Q8_0.gguf",
+            filename = "translator-lora-Q8_0.gguf",
+            compatibleModelIds = listOf("lfm2-350m-q4_k_m", "lfm2-350m-q8_0"),
+            fileSize = 690_176,
+            defaultScale = 1.0f,
+        ),
+        LoraAdapterCatalogEntry(
+            id = "sentiment-lora",
+            name = "Sentiment Analysis",
+            description = "Fine-tuned for sentiment analysis tasks",
+            downloadUrl = "https://huggingface.co/Void2377/Qwen/resolve/main/lora/sentiment-lora-Q8_0.gguf",
+            filename = "sentiment-lora-Q8_0.gguf",
+            compatibleModelIds = listOf("lfm2-350m-q4_k_m", "lfm2-350m-q8_0"),
+            fileSize = 690_176,
+            defaultScale = 1.0f,
+        ),
+        LoraAdapterCatalogEntry(
+            id = "uncensored-chat-lora",
+            name = "Uncensored Chat",
+            description = "Removes safety guardrails for uncensored responses",
+            downloadUrl = "https://huggingface.co/Void2377/Qwen/resolve/main/lora/uncensored_chat-lora-Q8_0.gguf",
+            filename = "uncensored_chat-lora-Q8_0.gguf",
+            compatibleModelIds = listOf("lfm2-350m-q4_k_m", "lfm2-350m-q8_0"),
+            fileSize = 1_372_160,
+            defaultScale = 1.0f,
+        ),
+    )
+
     // VLM
     private val vlmModels = listOf(
         AppModel(id = "smolvlm-500m-instruct-q8_0", name = "SmolVLM 500M Instruct",
@@ -146,6 +202,12 @@ object ModelList {
             }
         }
         Log.i(TAG, "VLM models registered (${vlmModels.size})")
+
+        // Register LoRA adapters
+        for (adapter in loraAdapters) {
+            RunAnywhere.registerLoraAdapter(adapter)
+        }
+        Log.i(TAG, "LoRA adapters registered (${loraAdapters.size})")
         Log.i(TAG, "All models registered")
     }
 }