Expose last-token candidates context method

* adds call to llama_get_logits, which accesses the
  result of the most recent llama_decode call
* "Token logits obtained from the last call to llama_decode()
   The logits for which llama_batch.logits[i] != 0
   are stored contiguously in the order they have
   appeared in the batch. Rows: number of tokens for
   which llama_batch.logits[i] != 0. Cols: n_vocab"
* https://github.com/ggerganov/llama.cpp/blob/c35e586ea57221844442c65a1172498c54971cb0/include/llama.h#L844-L849

Author: brittlewis12

llama-cpp-2/src/context.rs

@@ -187,6 +187,45 @@ impl<'model> LlamaContext<'model> {
         }
     }
 
+    /// Get the logits for the last token in the context.
+    ///
+    /// # Returns
+    /// An iterator over unsorted `LlamaTokenData` containing the
+    /// logits for the last token in the context.
+    ///
+    /// # Panics
+    ///
+    /// - underlying logits data is null
+    pub fn candidates(&self) -> impl Iterator<Item = LlamaTokenData> + '_ {
+        (0_i32..).zip(self.get_logits()).map(|(i, logit)| {
+            let token = LlamaToken::new(i);
+            LlamaTokenData::new(token, *logit, 0_f32)
+        })
+    }
+
+    /// Token logits obtained from the last call to `decode()`.
+    /// The logits for which `batch.logits[i] != 0` are stored contiguously
+    /// in the order they have appeared in the batch.
+    /// Rows: number of tokens for which `batch.logits[i] != 0`
+    /// Cols: `n_vocab`
+    ///
+    /// # Returns
+    ///
+    /// A slice containing the logits for the last decoded token.
+    /// The size corresponds to the `n_vocab` parameter of the context's model.
+    ///
+    /// # Panics
+    ///
+    /// - `n_vocab` does not fit into a usize
+    /// - token data returned is null
+    pub fn get_logits(&self) -> &[f32] {
+        let data = unsafe { llama_cpp_sys_2::llama_get_logits(self.context.as_ptr()) };
+        assert!(!data.is_null(), "logits data for last token is null");
+        let len = usize::try_from(self.model.n_vocab()).expect("n_vocab does not fit into a usize");
+
+        unsafe { slice::from_raw_parts(data, len) }
+    }
+
     /// Get the logits for the ith token in the context.
     ///
     /// # Panics