Add OpenAI/LiteLLM provider support with settings UI

- Add openai/ provider module with translator, client, convert, request/response translators
- Add shared provider/ types (ConversationMessage, MessageRole, ProviderConfig enum)
- Wire OpenAI-compatible provider dispatch alongside Bedrock in response_stream.rs
- Add ai.openai.* settings (enabled, base_url, api_key, model, models)
- Add OpenAI/LiteLLM settings page with model fetch, picker, and config UI
- Extend model menu items and llms.rs to surface LiteLLM models
- Update WARP.md with OpenAI provider architecture docs

app/src/ai/provider/mod.rs

@@ -0,0 +1,10 @@
+pub mod types;
+
+use crate::ai::bedrock::client::BedrockClientConfig;
+use crate::ai::openai::client::OpenAIClientConfig;
+
+pub enum ProviderConfig {
+    Bedrock(BedrockClientConfig),
+    OpenAI(OpenAIClientConfig),
+    None,
+}

app/src/ai/provider/types.rs

@@ -0,0 +1,51 @@
+use serde_json::Value as JsonValue;
+
+#[derive(Clone, Debug)]
+pub struct ConversationMessage {
+    pub role: MessageRole,
+    pub content: MessageContent,
+}
+
+#[derive(Clone, Debug, PartialEq)]
+pub enum MessageRole {
+    User,
+    Assistant,
+}
+
+#[derive(Clone, Debug)]
+pub enum MessageContent {
+    Text(String),
+    ToolUse {
+        tool_use_id: String,
+        name: String,
+        input: JsonValue,
+    },
+    ToolResult {
+        tool_use_id: String,
+        content: String,
+        is_error: bool,
+    },
+    MultiPart(Vec<ContentPart>),
+}
+
+#[derive(Clone, Debug)]
+pub enum ContentPart {
+    Text(String),
+    ToolUse {
+        tool_use_id: String,
+        name: String,
+        input: JsonValue,
+    },
+    ToolResult {
+        tool_use_id: String,
+        content: String,
+        is_error: bool,
+    },
+}
+
+#[derive(Clone, Debug)]
+pub struct ToolDefinition {
+    pub name: String,
+    pub description: String,
+    pub input_schema: JsonValue,
+}