Which standard deviation should I quote?
The sample one (s) in nearly every real situation. You use the population figure only when your list is the entire population, for instance the scores of all 12 members of a team when the team is the whole thing you care about. If your numbers are a sample that stands in for something bigger, dividing by n - 1 (Bessel's correction) is what removes the systematic under-estimate that dividing by n would give.
What happens to text or blank entries in my list?
They are skipped, counted, and shown back to you: the panel says how many non-numeric tokens were ignored and lists the first eight of them. That is deliberate, so that a stray header row or a currency symbol does not quietly change your mean. Check that line before you trust the numbers.
Why is there no mode for my list?
Because no value repeats. The mode is defined as the most frequent value, and when every value appears exactly once there is no meaningful answer, so the tool says "No mode (all unique)" rather than inventing one. If two or more values tie for the highest frequency, all of them are listed, since a list can genuinely be bimodal.
Can I compute a weighted average?
No. Every number counts once. If you need a weighted mean (for example a grade average where the exam counts triple), either repeat the value the right number of times in the list, which works when the weights are whole numbers, or compute it yourself as the sum of value times weight divided by the sum of the weights.
How precise are the results?
Values are shown to six significant figures and computed in standard 64-bit floating point. That is far more precision than any real measurement carries, so do not read meaning into the last digits; round to the precision of your input data when you report the result.