DateFacet suggestions are incredibly slow against large tables

[simonw]

I thought the limit 100 would help here, but it turns out that for a column on a huge table (e.g. 20m rows) which has no dates in at all that glob has to scan through every value of the column every time, just to figure out that there are NOT 100 matches!

datasette/datasette/facets.py

Lines 480 to 487 in f28ff8e

	# Does this column contain any dates in the first 100 rows?
	suggested_facet_sql = """
	select date({column}) from (
	{sql}
	) where {column} glob "????-??-*" limit 100;
	""".format(
	column=escape_sqlite(column), sql=self.sql
	)