Add TimerOutputs-based timers for internal kernels

[lkdvos]

Instruments the index manipulation, contraction and factorization kernels with @timeit_debug sections, compiled away by default and enabled via TensorKit.enable_timers!().

Section labels carry a category prefix (symmetry: / bookkeeping: / alloc: / dense:), and TensorKit.timer_summary() aggregates exclusive times into per-category totals that sum to the total measured time. The @cached macro times lookups separately from miss-path construction. While timing, get_num_*_threads() return 1 so that the (single-task) timer is never touched concurrently; the guard const-folds away when disabled, leaving production paths unchanged (verified via benchmark against main).

Adds TimerOutputs (1.x, for recursive enable_debug_timings) as a dependency, a "Profiling and timers" manual page, and tests.

🤖 Generated with Claude Code

[lkdvos]

Example output for a single (cold-cache) permute + svd_compact on an SU2Space(0 => 4, 1//2 => 4, 1 => 2) rank-4 tensor:

TensorKit.timer_summary()

    symmetry:  385μs ( 20.6%)   548KiB  39 sections
 bookkeeping:  195μs ( 10.5%)   159KiB  137 sections
       alloc: 25.4μs (  1.4%)   134KiB  3 sections
       dense: 1.19ms ( 63.8%)   228KiB  58 sections
       other: 70.7μs (  3.8%)  25.7KiB  3 sections

TensorKit.print_timers() — full call tree

 Section                                                  ncalls    time    %tot     avg
───────────────────────────────────────────────────────────────────────────────────────────
 svd_compact!                                                  1  1.20ms   64.2%  1.20ms
 ├─ dense: lapack                                              5  1.11ms   59.7%   223μs
 ├─ bookkeeping: cache sectorstructure                        24  19.8μs    1.1%   825ns
 │  └─ bookkeeping: compute sectorstructure                    1  2.15μs    0.1%  2.15μs
 └─ bookkeeping: cache degeneracystructure                    15  7.87μs    0.4%   525ns
 permute!                                                      1   579μs   31.0%   579μs
 └─ braid!                                                     1   579μs   31.0%   579μs
    ├─ bookkeeping: cache treebraider                          1   488μs   26.1%   488μs
    │  └─ symmetry: compute treebraider                        1   481μs   25.8%   481μs
    │     ├─ symmetry: recoupling matrices                     1   420μs   22.5%   420μs
    │     │  ├─ bookkeeping: cache fsbraid                    37   371μs   19.9%  10.0μs
    │     │  │  └─ symmetry: compute fsbraid                  37   347μs   18.6%  9.37μs
    │     │  └─ bookkeeping: repack                           37  16.0μs    0.9%   433ns
    │     ├─ bookkeeping: fusionblocks                         1  37.7μs    2.0%  37.7μs
    │     ├─ bookkeeping: cache sectorstructure                2  7.66μs    0.4%  3.83μs
    │     ├─ bookkeeping: cache degeneracystructure            2  6.41μs    0.3%  3.21μs
    │     └─ bookkeeping: sort                                 1  4.50μs    0.2%  4.50μs
    ├─ dense: tensoradd                                       29  36.9μs    2.0%  1.27μs
    ├─ dense: pack                                             8  21.4μs    1.1%  2.67μs
    ├─ dense: unpack                                           8  12.2μs    0.7%  1.53μs
    ├─ dense: recouple mul!                                    8  6.00μs    0.3%   750ns
    └─ alloc: buffers                                          1   735ns    0.0%   735ns
 alloc: initialize_output                                      1  50.6μs    2.7%  50.6μs
 └─ bookkeeping: cache degeneracystructure                     2  33.6μs    1.8%  16.8μs
 bookkeeping: cache degeneracystructure                        1  29.6μs    1.6%  29.6μs
 alloc: copy_input                                             1  9.30μs    0.5%  9.30μs

(some deeply nested sectorstructure/degeneracystructure sub-branches elided)

[codecov]

Codecov Report

❌ Patch coverage is 98.93617% with 3 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/planar/planaroperations.jl	96.29%	1 Missing ⚠️
src/tensors/indexmanipulations.jl	97.50%	1 Missing ⚠️
src/tensors/treetransformers.jl	92.30%	1 Missing ⚠️

Files with missing lines	Coverage Δ
src/TensorKit.jl	17.24% <100.00%> (ø)
src/auxiliary/caches.jl	89.88% <100.00%> (+0.35%)	⬆️
src/auxiliary/timers.jl	100.00% <100.00%> (ø)
src/factorizations/factorizations.jl	84.61% <ø> (ø)
src/factorizations/matrixalgebrakit.jl	93.83% <100.00%> (+1.09%)	⬆️
src/tensors/linalg.jl	82.93% <100.00%> (+0.05%)	⬆️
src/tensors/tensoroperations.jl	96.55% <100.00%> (-0.83%)	⬇️
src/planar/planaroperations.jl	72.46% <96.29%> (-0.34%)	⬇️
src/tensors/indexmanipulations.jl	89.14% <97.50%> (-1.30%)	⬇️
src/tensors/treetransformers.jl	94.84% <92.30%> (-0.95%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.