Cleanup and optimize render_impls

[yotamofek]

• take ownership of the Vec<&Impl> instead of copying into another alloc
• reuse ImplString to do natural sort ordering
• lazy formatting

Somewhat of a follow-up to #157233 and #157179 (cc @nnethercote - thanks!)
This kinda undoes f7c8bc2 but IMHO it makes more sense to be explicit about negative impl ordering, and also seems kinda wasteful to "render" the negativity into a string and rely on however ASCII decided to order characters. I can also undo this part, I think this PR is still a positive change even without it.

r? @GuillaumeGomez

LLM disclosure:
I used LLM for reviewing my changes and making sure some assumptions I was making were correct (i.e. that render_impls will not render anything IFF the list of traits passed to it is empty)

[yotamofek]

I doubt it'll register on perf, but I do think this should be an optimization (albeit a small one) so let's run perf anyways

@bors try @rust-timer queue

[yotamofek]

A job failed! Check out the build log: (web) (plain enhanced) (plain)
Click to see the possible cause of the failure (guessed by this bot)

Docker error
hopefully try build will still work

[rust-bors]

☀️ Try build successful (CI)
Build commit: 19872c0 (19872c0f5e198bd2f48c9898b0b0e83f77aa8dbf, parent: 8954863c81df429ebf96ea38a16c76f209995833)

[rust-timer]

Finished benchmarking commit (19872c0): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

@bors rollup=never
@rustbot label: -S-waiting-on-perf -perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.2%	[-0.5%, -0.1%]	8
Improvements ✅ (secondary)	-0.2%	[-0.3%, -0.2%]	3
All ❌✅ (primary)	-0.2%	[-0.5%, -0.1%]	8

Max RSS (memory usage)

Results (primary 0.6%, secondary 2.7%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	7.0%	[6.8%, 7.1%]	2
Regressions ❌ (secondary)	3.9%	[2.6%, 6.0%]	4
Improvements ✅ (primary)	-3.6%	[-5.2%, -2.6%]	3
Improvements ✅ (secondary)	-2.0%	[-2.0%, -2.0%]	1
All ❌✅ (primary)	0.6%	[-5.2%, 7.1%]	5

Cycles

Results (primary 3.4%, secondary 3.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	3.4%	[1.5%, 6.0%]	102
Regressions ❌ (secondary)	3.5%	[2.1%, 7.0%]	90
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.4%	[1.5%, 6.0%]	102

Binary size

Results (primary 0.0%, secondary 0.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	0.0%	[0.0%, 0.1%]	30
Regressions ❌ (secondary)	0.0%	[0.0%, 0.1%]	6
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.0%	[0.0%, 0.1%]	30

Bootstrap: 514.086s -> 513.359s (-0.14%)
Artifact size: 400.74 MiB -> 400.64 MiB (-0.03%)

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[yotamofek]

Perf results are a pleasant surprise! 🦆
Force pushed without any changes to trigger new CI run

[GuillaumeGomez]

Nice, well done! r=me once CI pass.

[rust-log-analyzer]

The job aarch64-gnu-llvm-21-1 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

---- [rustdoc-html] tests/rustdoc-html/macro/const-rendering-macros-33302.rs stdout ----
------python3 stdout------------------------------

------python3 stderr------------------------------
29: has check failed
 `XPATH PATTERN` did not match
         //@ has - '//*[@id="associatedconstant.C"]' 'const C: [i32; 16]'
45: has check failed
 `XPATH PATTERN` did not match
         //@ has - '//*[@id="associatedconstant.C-2"]' 'const C: (i32, i32)'

Encountered 2 errors

------------------------------------------

error: htmldocck failed!
status: exit status: 1
command: "/usr/bin/python3" "/checkout/src/etc/htmldocck.py" "/checkout/obj/build/aarch64-unknown-linux-gnu/test/rustdoc-html/macro/const-rendering-macros-33302" "/checkout/tests/rustdoc-html/macro/const-rendering-macros-33302.rs"
stdout: none
--- stderr -------------------------------
29: has check failed
 `XPATH PATTERN` did not match
         //@ has - '//*[@id="associatedconstant.C"]' 'const C: [i32; 16]'
45: has check failed
 `XPATH PATTERN` did not match
         //@ has - '//*[@id="associatedconstant.C-2"]' 'const C: (i32, i32)'

Encountered 2 errors
------------------------------------------

---- [rustdoc-html] tests/rustdoc-html/macro/const-rendering-macros-33302.rs stdout end ----