docs: add missing notes

lib/node_modules/@stdlib/fft/base/fftpack/ndarray/generic/cffti/README.md

@@ -81,6 +81,30 @@ The function has the following parameters:
 
 ## Notes
 
+-   Let `N` equal the number of elements in the input ndarray. The input ndarray is divided into three sections:
+
+    ```text
+              size = 2N                2N            2+ceil(log2(N)/2)
+                   ↓                   ↓                  ↓
+        | scratch / workspace | twiddle factors | radix factor table |
+                   ↑                   ↑                  ↑
+    i = 0         ...         2N       ...       4N       ...
+    ```
+
+    -   **scratch/workspace**: used as a scratch space when performing transforms. This section is not updated during initialization.
+    -   **twiddle factors**: a table of reusable complex-exponential constants stored as cosine/sine pairs.
+    -   **radix factor table**: a table containing the sequence length `N`, the number of factors into which `N` was decomposed, and the individual integer radix factors.
+
+-   In general, an input ndarray should have `4N + 34` indexed elements (as `log2(N)/2 ≤ 32` for all `2^64`). During initialization, only the sections for storing twiddle factors and the factorization of `N` are updated.
+
+-   The radix factor table is comprised as follows:
+
+    ```text
+    | sequence_length | number_of_factors | integer_factors |
+    ```
+
+-   If `N` equals `1`, the function returns early without modifying the input ndarray, as a single data point is its own Fourier transform.
+
 -   If provided an empty one-dimensional input ndarray, the function returns the input ndarray unchanged.
 
 </section>

lib/node_modules/@stdlib/fft/base/fftpack/ndarray/generic/rffti/README.md

@@ -81,6 +81,30 @@ The function has the following parameters:
 
 ## Notes
 
+-   Let `N` equal the number of elements in the input ndarray. The input ndarray is divided into three sections:
+
+    ```text
+              size = N                 N            2+ceil(log2(N)/2)
+                   ↓                   ↓                  ↓
+        | scratch / workspace | twiddle factors | radix factor table |
+                   ↑                   ↑                  ↑
+    i = 0         ...         N       ...       2N       ...
+    ```
+
+    -   **scratch/workspace**: used as a scratch space when performing transforms. This section is not updated during initialization.
+    -   **twiddle factors**: a table of reusable complex-exponential constants stored as cosine/sine pairs.
+    -   **radix factor table**: a table containing the sequence length `N`, the number of factors into which `N` was decomposed, and the individual integer radix factors.
+
+-   In general, an input ndarray should have `2N + 34` indexed elements (as `log2(N)/2 ≤ 32` for all `2^64`). During initialization, only the sections for storing twiddle factors and the factorization of `N` are updated.
+
+-   The radix factor table is comprised as follows:
+
+    ```text
+    | sequence_length | number_of_factors | integer_factors |
+    ```
+
+-   If `N` equals `1`, the function returns early without modifying the input ndarray, as a single data point is its own Fourier transform.
+
 -   If provided an empty one-dimensional input ndarray, the function returns the input ndarray unchanged.
 
 </section>