Please address the comments from this code review:
## Overall Comments
- Consider deriving the servo column names dynamically from self.servos instead of hard-coding "Servo1"–"Servo4", so the methods will work with any number of servos.
- In export_timeline_to_csv, validate or create the output folder if it doesn’t exist (and surface an error if it can’t be written), and skip the timestamp column on import rather than assuming fixed position.

## Individual Comments

### Comment 1
<location> `pslab/external/motor.py:136` </location>
<code_context>
+            reader = csv.DictReader(csvfile)
+            for row in reader:
+                angles = []
+                for key in ["Servo1", "Servo2", "Servo3", "Servo4"]:
+                    value = row.get(key, "").strip().lower()
+                    if value in ("", "null", "none"):
+                        angles.append(None)
+                    else:
+                        angles.append(int(value))
+                timeline.append(angles)
+
</code_context>

<issue_to_address>
Assumes all servo columns are present and in order.

If columns are missing or out of order, parsing may fail or produce incorrect results. Please validate the header or handle columns more flexibly.
</issue_to_address>

### Comment 2
<location> `pslab/external/motor.py:166` </location>
<code_context>
+
+        with open(filepath, mode="w", newline="") as csvfile:
+            writer = csv.writer(csvfile)
+            writer.writerow(["Timestamp", "Servo1", "Servo2", "Servo3", "Servo4"])
+            for i, row in enumerate(timeline):
+                pos = ["null" if val is None else val for val in row]
</code_context>

<issue_to_address>
Header uses 'Timestamp' for what is actually a timestep index.

Consider renaming the column to 'Timestep' or 'Step' to better reflect its contents.
</issue_to_address>

<suggested_fix>
<<<<<<< SEARCH
            writer.writerow(["Timestamp", "Servo1", "Servo2", "Servo3", "Servo4"])
=======
            writer.writerow(["Timestep", "Servo1", "Servo2", "Servo3", "Servo4"])
>>>>>>> REPLACE

</suggested_fix>

### Comment 3
<location> `pslab/external/motor.py:168` </location>
<code_context>
+            writer = csv.writer(csvfile)
+            writer.writerow(["Timestamp", "Servo1", "Servo2", "Servo3", "Servo4"])
+            for i, row in enumerate(timeline):
+                pos = ["null" if val is None else val for val in row]
+                writer.writerow([i] + pos)
</code_context>

<issue_to_address>
Inconsistent handling of None values between import and export.

Consider using a single representation for None values (e.g., always 'null') in both import and export, and document this behavior for clarity.

Suggested implementation:

```python
        # Export timeline to CSV, using 'null' to represent None values.
        # When importing, 'null' will be interpreted as None.
        with open(filepath, mode="w", newline="") as csvfile:
            writer = csv.writer(csvfile)
            writer.writerow(["Timestamp", "Servo1", "Servo2", "Servo3", "Servo4"])
            for i, row in enumerate(timeline):
                pos = ["null" if val is None else val for val in row]
                writer.writerow([i] + pos)

```

```python
def import_timeline_from_csv(filepath):
    """
    Import a timeline from a CSV file, interpreting 'null' as None.

    Args:
        filepath (str): Path to the CSV file.

    Returns:
        list: Timeline data as a list of lists, with None for 'null' values.
    """
    timeline = []
    with open(filepath, mode="r", newline="") as csvfile:
        reader = csv.reader(csvfile)
        headers = next(reader)  # Skip header
        for row in reader:
            # row[0] is the timestamp/index, skip or store as needed
            positions = [None if val == "null" else float(val) for val in row[1:]]
            timeline.append(positions)
    return timeline

```

- If there is already an import function, update its logic to use `None if val == "null" else ...` for consistency.
- Ensure that all documentation and usage of import/export functions mention that 'null' is the standard representation for missing values.
- If the import function is used elsewhere, update those usages to use the new or updated function.
</issue_to_address>