BW
The amount of work you must have done to dig up all that deserves a better response than you are going to get here
The answer depends on where one is going.
If going around the SE, mostly under the LTMA, then I would scrub all of them. The chance is that one would get away with it 90% of the time, because most of the time, in non-cumulus cloud, the ice buildup is between zero and thin. But I wouldn't take the chance without an escape route, and that would be diverting somewhere completely different.
If going to south of France where the OAT is say 10C higher, and the weather over France was either clear or I was sure of being able to stay VMC on top, then I would depart in all of them. With a de-iced prop, there is no problem in climbing 2000ft through cloud especially if scattered/broken.
If going from the SE to say the Isle of Man then again I would scrub all of them because the conditions will get only worse up north, and one can't go high enough until later in the flight. The exception would be excellent conditions (actual and forecast) soon into the flight in which case it would be worth going around the LTMA.
But you have an unfair advantage: you have cloud tops data. In most of the examples you have, the tops cannot be judged from the ground. This is the big missing item from presently available weather data: a picture of the tops en-route. Bottoms are easy, from TAFs/METARs. The other bit is the 0C level - Form 214/215 is too rough.
Do I need to attend a weather course?
Come to think of it, I've been looking for one for ages and nobody does anything useful. Lots of basic PPL VFR stuff... What is needed is a specialised course which tells you how to get the data out of GFS and anywhere else and interpret it into an en-route cloud/temperature profile.