how to find the correlation between categorical and numerical columns

I would check this by plotting the data. The colors have to be somewhat horizontally

import seaborn as sns
sns.scatterplot(data=df, x=df.index, y='Light_Sensor_Reading', hue='Light_Sensor_Status')

Correlation here feels a bit wrong, but I guess you could factorize the status.

Assuming you have your data in a Pandas DataFrame df:

from pandas import factorize

labels, categories = factorize(df["Light_Sensor_Status"])
df["labels"] = labels
abs(df["Light_Sensor_Reading"].corr(df["labels"]))

Correlation should give you a value of 0.9998202098949396.

I encode the status as number then use the dataframe pearson correlation function to return the correlation

from sklearn.preprocessing import LabelEncoder
Light_Sensor_Reading=[231, 231,231,231,231,231,231,231,231,232,950,988,987,986,986,986,986,986,986,986,986,986,986,985,985]
Light_Sensor_Status=['Dim','Dim','Dim','Dim','Dim','Dim','Dim','Dim','Dim','Dim','Very_bright','Very_bright','Very_bright', 'Very_bright',   'Very_bright', 'Very_bright','Very_bright','Very_bright','Very_bright','Very_bright','Very_bright',\
'Very_bright','Very_bright','Very_bright','Very_bright']

df=pd.DataFrame({'light_sensor_reading': Light_Sensor_Reading,'light_sensor_status':Light_Sensor_Status})
x=range(len(df))
sns.scatterplot(x=x, y='light_sensor_reading',data=df,hue='light_sensor_status')
plt.show()
encoder=LabelEncoder()
encoder.fit(df['light_sensor_status'])
df['light_sensor_status_enc']=encoder.transform(df['light_sensor_status'])
corr=df['light_sensor_reading'].corr(df['light_sensor_status_enc'])
print("correlation ",corr)

output:

correlation  0.9998202098949396

it doesn't mean anything to calculate the correlation between two variables if they are not quantitative.

I think what you want to do is to study the link between them.

The purpose is to explain the first variable with the other one through a model.

You can use the logistic regression.

But you will only know if there is a link between these two variables with a T-Test or a wilcoxon test depending on the normality of the data