- Too few spells known. The smallest amount out of all full casters. This makes the class harder to play than it should be, and limits the ability to cover a theme because there are just not enough spells.
- Limited class spells. A lot of the sorcerer class spells focus on combat and there aren't enough non-combat utility effects. Also there are basically no exclusive/sorcerer only spells. And most egregious, Dragon sorcerer can't turn into a dragon despite the spell existing in the game!
- Too much pressure on sorcery points. Sorcerer features rely too much on sorcery points. There are too many uses for sorcery points and not enough of them to cast even a full day's assortment of spells and use metamagic, let alone use the other class features.
- Limited Metamagic. Metamagic takes too long to come online and once you have it you have too few metamagic options for most of your career. Some options are nearly must-haves and some are too expensive for what they do. Some options aren't in the core book (like damage change)
- Issues with subclasses. Almost all older subclasses have issues. Wild magic is too reliant on DM cooperation, Dragon magic works fine, as long as you choose a fire dragon... most older subclasses could benefit from bonus spells, and there are only two in core.
- No familiars.
Some minor nitpicks that are more personal:
- Lack of simple weapon proficiency. Spears are an iconic sorcerer weapon, at least let us have proficiency with them. More of a ribbon since most everybody will just blast, but come on! 5e took them away for no reason!
- Needing an arcane focus or material components. It would be nice, real nice if sorcerers could cast without needing to rely on external items. Bring Eschew Materials back!
- Some more nice to have stuff. Like I don't know letting sorcerers have an always-on weaker version of detect magic.
I'll be the first to admit that I don't like the Sorcerer class, but since I know it isn't going away... I think most of the things you have highlighted are all reasonable and doable in a Sorcerer revision.
I think Sorcerers gaining more spells known... or the other method, which is each subclass gains a pair of subclass spells each spell level that don't count against their spells known total is completely reasonable and I imagine will be done.
I think most metamagics are underpowered enough that you don't need to worry about making more of them available to Sorcerer PCs. Personally, I think the easiest/most effective thing to do is have PCs select metamagics they "know" like they do right now... but that they can also use any other metamagics by spending an additional Sorcery Point over their normal cost.
I think Twinned Spell should be removed as a metmagic, as that and Quickened Spell are the two metamagics that cause the most hand-wringing in terms of "overpoweredness"... and it would just save time and energy to remove it entirely as an option and just not try and balance it across every single spell available currently to be Twinned. (Especially considering Sorcerers would be gaining all the other metamagics as an extra thing to replace it with if my point above was incorporated.)
I think there should be a really long thought and playtest of the idea of switching the Sorcerer over to the Spell Point variant, and adding Spell Points and Sorcery Points together into a single pool. That would solve the lack of Sorcery Points issue, and will make a genuine feeling of change and difference between Sorcerers and other spellcasters. But I do think for this to be workable it would need a crap-ton of playtesting... as Sorcerers would now have many more points to spend on metamagics and other abilities, while at the same time perhaps hamstringing themselves if they blow too many points on those things and thus not having enough left over for a proper amount of spells cast in a day. A Spell Point / Sorcery Point combo would open up many more options... but would be harder for players to guesstimate how best to spend the points in the course of a day.
As far as spell lists themselves... I'm torn. On the one hand, I am a firm believer that every single Sorcerer subclass should have a smaller but distinct spell list for just that subclass, with spells specifically chosen to apply to that subclass's theme. So no
Fireballs available to Storm Sorcerers for example, but the primal spell
Call Lightning is made available instead. But on the other hand... I also fully realize the amount of page real estate you'd have to add to the game to get that done, and that it really probably isn't feasible. Because you're talking every subclass having a spell list of 9 levels worth of spells, plus probably every color of Dragon sorcerer having their own too. That's I think just way too many pages to spend on something that (while cool) just doesn't affect enough people to warrant going that far. So in that regard, I think just sticking with the one Sorcerer spell list and then the additional two spells known from subclass as above is probably the way to go (which perhaps a sidebar in the PHB or DMG on how to create custom spell lists for individual subclasses.)
Familiars? I think on the one hand it's so inconsequential that there's no reason you couldn't add it to the Sorcerer spell list... but I also think it's so inconsequential to any one particular Sorcerer that not having it on the Sorcerer spell list and instead making it available via the Magic Initiate feat at 1st level for those players who REALLY want it would be fine too. If a Sorcerer player wants a familiar they can now get it... but making them available to every Sorcerer is not necessarily needed. So whatever the survey would say about wants/needs for Sorcerer familiars would be fine with me.
Simple Weapon proficiency? Fine. It's a ribbon as you say, so whatever.
Removal of material components? Add a metamagic for it and I'm good with that idea. Remove them
entirely from Sorcerers though? Like I've said in all the Psion class discussions... this is one that for balance reasons I completely understand why WotC would not make this change. The game is set up that all spellcasters need components to cast spells across the board. It's easy to remember and there's no questioning of balance or whatever. But the game also makes the ignoring or removing of material components so easy-- either by just using focuses which can be completely forgotten about after they've been bought-- or by letting individual DMs just choose not to use them for their Sorcerers and Psions if that matters to them. But no one needs the book itself to remove material components for them.
*****
In terms of other potential sorcerer abilities? Unfortunately I think the bigger issue with Sorcerers there compared to Wizards is that the Sorcerer's narrative of character really comes out of their subclass and
not their class. For Wizards... it doesn't matter what spell school you focus on, all the narrative parts of who you are is from the Wizard class itself. Regardless of school, you still treat magic like a science, you still have to memorize your spells via spell books, you still can acquire new spells by copying from or using other spell books as though they were instruction manuals, etc. But for Sorcerers... who you are comes directly out of your Origin. Your subclass gives us the narrative of how you acquired your ability to manipulate magic. And the term 'Sorcerer' is almost just a title heading for bringing all these individual casters together, as opposed to an actual thing you are. But that makes it harder to give the class itself more goodies, as those goodies have to be generic enough to apply to every single Origin out there. Like all Sorcerers being able to detect magic at will? Should an Aberrant Mind be able to do that? Does that make sense? Mayyyyyyyyybe? I'm sure someone could reason out why this
psionic character might detect
magic, but it doesn't automatically make sense. And that's going to be the true of a lot of potential "Sorcerer" features one might want to add to the overarching class. Clockwork Souls and Divine Souls and Wild Mages and Storm Sorcerers are all
very different animals... and thus finding features that would make sense to all of them will require a bit of doing. Not impossible by any stretch... but definitely would require some careful thought.