Fix: banner on launch, config file loading, 7 providers match setup wizard

BUGFIXES:
- Banner now displays on 'gyandoot run' via gyandoot_tui::banner::print_banner()
- Config now loads from ~/.config/gyandoot/config.toml with fallback chain
  (XDG -> ~/.gyandoot/config.yaml -> hardcoded defaults), fixing 401 errors
- gyandoot init now writes visible placeholder api_key for users to edit

NEW PROVIDERS (4 added, now 7 total matching setup wizard):
- Google Gemini: Native API with x-goog-api-key auth, streamingGenerateContent
- DeepSeek: OpenAI-compatible wrapper at api.deepseek.com/v1
- OpenRouter: With HTTP-Referer/X-Title headers, model routing
- Groq: OpenAI-compatible wrapper at api.groq.com/openai/v1

All compile with 0 errors.

crates/gyandoot-providers/src/groq.rs

@@ -0,0 +1,107 @@
+//! Groq API provider implementation.
+//!
+//! OpenAI-compatible chat completions endpoint (`/v1/chat/completions`) with
+//! Bearer token auth. Supports streaming and function calling identically
+//! to OpenAI. Base URL: `https://api.groq.com/openai/v1`.
+
+use async_trait::async_trait;
+use gyandoot_core::{ConversationContext, ProviderResult};
+use std::time::Duration;
+use tracing::debug;
+
+use crate::openai::OpenAIProvider;
+use crate::{Provider, ProviderResponse};
+
+const GROQ_BASE_URL: &str = "https://api.groq.com/openai/v1";
+
+/// Provider for the Groq API.
+///
+/// Groq uses an OpenAI-compatible chat completions endpoint.
+/// This implementation wraps [`OpenAIProvider`] with the correct base URL.
+pub struct GroqProvider {
+    inner: OpenAIProvider,
+}
+
+impl GroqProvider {
+    /// Create a new Groq provider.
+    ///
+    /// - `api_key`: Groq API key (sent as Bearer token).
+    /// - `model`: Model identifier (e.g. `"llama-3.3-70b-versatile"`).
+    /// - `base_url`: Optional override for the API base URL.
+    pub fn new(api_key: String, model: String, base_url: Option<String>) -> Self {
+        let base = base_url.unwrap_or_else(|| GROQ_BASE_URL.to_string());
+        Self {
+            inner: OpenAIProvider::new(api_key, model, Some(base)),
+        }
+    }
+
+    /// Set a custom request timeout.
+    pub fn with_timeout(mut self, timeout: Duration) -> Self {
+        self.inner = self.inner.with_timeout(timeout);
+        self
+    }
+}
+
+#[async_trait]
+impl Provider for GroqProvider {
+    async fn chat(&self, context: ConversationContext) -> ProviderResult<ProviderResponse> {
+        debug!(
+            "Groq chat request (model: {}, messages: {})",
+            self.inner.model,
+            context.messages.len()
+        );
+        self.inner.chat(context).await
+    }
+
+    async fn chat_stream(
+        &self,
+        context: ConversationContext,
+    ) -> ProviderResult<Box<dyn tokio_stream::Stream<Item = ProviderResult<String>> + Send + Unpin>>
+    {
+        debug!(
+            "Groq streaming chat (model: {}, messages: {})",
+            self.inner.model,
+            context.messages.len()
+        );
+        self.inner.chat_stream(context).await
+    }
+
+    async fn models(&self) -> ProviderResult<Vec<String>> {
+        Ok(vec![
+            "llama-3.3-70b-versatile".to_string(),
+            "llama-3.1-8b-instant".to_string(),
+            "mixtral-8x7b-32768".to_string(),
+            "gemma2-9b-it".to_string(),
+            "deepseek-r1-distill-llama-70b".to_string(),
+        ])
+    }
+
+    fn supports_tools(&self) -> bool {
+        true
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    fn test_provider() -> GroqProvider {
+        GroqProvider::new(
+            "gsk-test".to_string(),
+            "llama-3.3-70b-versatile".to_string(),
+            None,
+        )
+    }
+
+    #[test]
+    fn test_creation() {
+        let provider = test_provider();
+        assert_eq!(provider.inner.model, "llama-3.3-70b-versatile");
+    }
+
+    #[test]
+    fn test_supports_tools() {
+        let provider = test_provider();
+        assert!(provider.supports_tools());
+    }
+}