fix(seer): Ensure count is a valid float in TimeSeriesPoint (#91809)

👋 Hi there! This PR was automatically generated by Autofix 🤖

This fix was triggered by Rohan Agarwal.

Fixes
[SEER-2N](https://sentry.io/organizations/sentry/issues/5028696400/).
The issue was that: `format_crash_free_data` passes Snuba `None` values
as `TimeSeriesPoint.value`, failing Seer's Pydantic float validation.

- Ensures that the `count` value used to create `TimeSeriesPoint` is
always a valid float.
- Handles cases where Snuba returns `None` for the count by converting
it to `0.0`.

If you have any questions or feedback for the Sentry team about this
fix, please email [autofix@sentry.io](mailto:autofix@sentry.io) with the
Run ID: 44967.

Co-authored-by: sentry-autofix[bot] <157164994+sentry-autofix[bot]@users.noreply.github.com>

Author: seer-by-sentry[bot]

src/sentry/seer/anomaly_detection/utils.py

@@ -157,7 +157,8 @@ def format_crash_free_data(data: SnubaTSResult) -> list[TimeSeriesPoint]:
 
     for time, count in zip(nested_data.get("intervals"), series.get("sum(session)", 0)):
         date = datetime.strptime(time, "%Y-%m-%dT%H:%M:%SZ")
-        ts_point = TimeSeriesPoint(timestamp=date.timestamp(), value=count)
+        count_value = 0.0 if count is None else float(count)
+        ts_point = TimeSeriesPoint(timestamp=date.timestamp(), value=count_value)
         formatted_data.append(ts_point)
     return formatted_data