Add FPS (Formal Power Series) benchmark workload

[shakayami]

Summary

This PR adds a new benchmark workload for Formal Power Series (FPS) operations to the benchmarks suite.

Key Changes

• Import the fps module at the top of the workloads file
• Add build_fps_workload() function that generates a random sequence of n=8,192 elements modulo 998244353
• Add run_fps() function that performs FPS operations: exponential, logarithm, and resize operations
• Register the new FPS benchmark in the BENCHMARKS list

Implementation Details

The FPS workload:

• Generates a sequence of 8,192 random integers (with the first element fixed at 0)
• Tests three key FPS operations: exp() (exponential), log() (logarithm), and resize() (resizing to original length)
• Uses the same modulo constant (998244353) as other benchmarks for consistency

https://claude.ai/code/session_011dKfP4nUzhqoGypWQNJxta

[github-actions]

📊 Performance Benchmark Results

⏱️ Execution time (mean, lower is better)

Benchmark	Base	PR	Delta
convolution	959.67ms	884.09ms	-7.9% ✅
dsu	105.99ms	106.49ms	+0.5%
fps	n/a	570.37ms	n/a
lazysegtree	1.631s	1.665s	+2.0%
maxflow	11.24ms	12.49ms	+11.1% ⚠️
mincostflow	5.13ms	4.67ms	-9.0% ✅
scc	146.60ms	144.72ms	-1.3% ✅
segtree	245.07ms	246.32ms	+0.5%

💾 Peak memory (lower is better)

Benchmark	Base	PR	Delta
convolution	16.1MiB	16.1MiB	+0.0%
dsu	849.2KiB	849.2KiB	+0.0%
fps	n/a	3.9MiB	n/a
lazysegtree	5.0MiB	5.0MiB	+0.0%
maxflow	2.3MiB	2.3MiB	+0.0%
mincostflow	662.2KiB	662.2KiB	+0.0%
scc	13.8MiB	13.8MiB	+0.0%
segtree	1.5MiB	1.5MiB	+0.0%

_{Base = target branch, PR = this branch. Negative delta = improvement. ✅ improvement ≥ 1%, ⚠️ regression > 5%. "n/a" base entries mean the target branch has no benchmark suite yet.}