So, there is definitely a systemic problem here, that is impossible to fix and difficult to work around, as Theik says, and it’s certainly true that E6 will help keep things from getting worse, as kviiri says, because things are very much liable to get worse at 7th and beyond.

But that’s not the whole story.

I think you have systemic problems compounded by player problems. To wit, the stronger classes (read: easier to get working well, requiring less optimization skill) are being played by those optimizing more, rather than those who are optimizing less, which might help to balance out the situation. In particular, Divine Metamagic is probably much-too-high-power for a game in which the barbarian and fighter aren’t optimizing much and falling into many of the myriad traps before those classes.

And on the flip side, the weaker classes (read: harder to get working well, requiring greater optimization skill) are being played by those optimizing less, rather than those who were optimizing more. It’s counter-intuitive, since the weaker classes are also branded the “simpler” ones, but they are actually harder to play. There are just so many traps for a fighter or a barbarian.

Which they must have fallen into, because a simple summon monster with Augment Summoning shouldn’t be enough for a summoned creature to be superior to a barbarian or a fighter. Are there options out there for a wizard to summon things so powerful that nothing a barbarian or fighter can do will help? Yes, there are—but they’re more involved than just the one feat.

For the bard, there is probably plenty of room for improvement just in feats and spells. This Q&A might be of assistance there. The bard isn’t cleric or wizard, but they can still definitely contribute.

The barbarian, fighter, and paladin are going to have a harder time.

So what I would recommend here is to consider suggesting that the barbarian, fighter, and paladin play better (read: easier to get working well) classes. Specifically, Tome of Battle was something of a revolution in D&D 3.5e design, finally working hard to ensure that there were no traps and things worked as well as they sounded on paper, giving martial characters nice things that enabled them to fight as well as mages cast (well, almost), and they’re pretty newbie-friendly.

Specifically, in my games, I have encouraged players’ class choices as follows:

• fighter → warblade,
• paladin → crusader, and
• barbarian → also crusader, but with the Devoted Spirit discipline swapped for the Iron Heart discipline
  • (and, if desired, the White Raven discipline swapped for the Tiger Claw discipline, though White Raven can work quite well for a barbarian and Tiger Claw is mostly only useful if you’re interested in dual-wielding).

These classes match the fluff and narrative role of these classes very well, to the point that they’re often considered just replacements for those classes (and third class in the book, swordsage, is a replacement for the monk or ninja classes). So what I recommend is that your 5th-level paladin, fighter, and barbarian become 5th-level crusader, warblade, and “crusader,” instead. The characters can stay the same, and gain substantially improved combat ability.

Tome of Battle also multiclasses very well—5 levels of barbarian, fighter, or paladin is hard to countenance, but the first 2 levels of each of those classes is quite good. That allows the barbarian to keep his iconic rage, the fighter to keep his armor and shield proficiencies as well as all but one of his bonus feats, and the paladin to keep detect evil, smite evil, and divine grace. Levels of non-Tome of Battle classes count half towards Tome of Battle “initiator level” (similar to caster level), so those levels also help their maneuvers (I have also had success with allowing fighter levels to count full for initiator level; fighter is a weaker class than barbarian or paladin). So these characters, if they did keep 2 levels in their original classes, would start with IL 2nd, get IL 3rd at their 2nd class level, and so on, allowing them to choose higher-level maneuvers sooner.

You can actually go further and take 4 non-initiating levels, so you start at IL 3rd and can select 2nd-level maneuvers and stances with your initial set of maneuvers, but this is minor and the three classes we’re discussing don’t get a whole lot at 3rd and 4th (the paladin’s doing a bit better than the others, and grabbing the aura of courage and turn undead might be worthwhile—and then they could go for the ruby knight vindicator prestige class, adapted as necessary, in the same book).

So I think if your party had a 2nd-level barbarian/3rd-level crusader, a 2nd-level fighter/3rd-level warblade (or just a 5th-level warblade, honestly), and a 2nd-level paladin/3rd-level crusader (or 4th-level paladin/1st-level crusader or 5th-level crusader or whatever), you would likely be in a much better place at 5th level than you are now.

The spellcasters should not have a lot of trouble with this; spellcasters are still more powerful classes. And they’ll still get (a lot) more powerful still at 7th level and beyond—E6 would still be a good idea. If you do that, I would lift the 3× restriction on Martial Study, and definitely suggest to the paladin that taking more than 2 levels of paladin (and thus never hitting IL 5th in 6 levels) would be very costly (and I might allow some feat to “catch up,” à la Practiced Spellcaster, though I would have to think carefully about what would be appropriate there because that might make their aura of courage and turn undead seem rather like freebies).

Note, Tome of Battle doesn’t spell it out and I think it should: the way to handle crusader maneuvers is with a little deck of cards. Wizards of the Coast made a free set you can print, if you want, or you can just pull scraps of paper out of a hat or just grab some playing cards and write down that the Ace of Spades is your mountain hammer or whatever. Point is, the crusader has some randomness with their maneuvers, and dice are the wrong tool to use to figure them out. With cards, you just draw cards as you get maneuvers, and then when the deck’s run out, you reshuffle the deck and start over: easy. With dice, you have to worry about odd numbers, and then which number is which maneuver discounting those already granted, and it’s just not worth it—cards are the way to go.

And if you do print out the maneuver cards, those have the extra advantage of having the full rules text of every maneuver right there in front of the players. They’ll know exactly what they can do and how it works, physically on the table in front of them. This is great for players who are daunted by the idea of resource management and special powers. These advantages, in fact, apply to all initiators, not just crusaders—swordsages and warblades never have to shuffle, but they certainly can keep cards in front of them to remind themselves what they have readied, and flip them over as they use them.

Not all weapons are equally powerful and they don't need to be

The light hammer is not the only weapon that you would never use from a perspective of power.

The sickle is equally bad as the light hammer. Similarly, there is really no reason to specifically use a trident. It is the same as a spear but needs martial weapons proficiency.

When you start to consider that every weapon should have their advantages, you have even more work.

Then again, I don't see why it is absolutely necessary to only have weapons that have a use for the optimizer.

Finesse seems very non-thematic on a hammer

While adding finesse to the light hammer won't have many mechanical consequences, it means that Dex based characters now have a finesse bludgeoning weapon against skeletons. It is true that this probably won't come up that often and you said you don't find this important but it is a fact. It won't break your game.

On the other hand, as stated above, I don't think there is a problem to be fixed. And adding finesse to a hammer seems very non-thematic.