rubocop: reduce redundant block.call for performance

[kenhys]

Which issue(s) this PR fixes:
Fixes #

What this PR does / why we need it:

This is cosmetic change, it does not change behavior at all. The following rubocop configuration detects it.

Performance/RedundantBlockCall:
  Enable: true

Benchmark result:

ruby 3.2.8 (2025-03-26 revision 13f495dc2c) +YJIT [x86_64-linux]
Warming up --------------------------------------
          block call     1.296M i/100ms
   yield with &block     1.708M i/100ms
yield without &block     1.687M i/100ms
Calculating -------------------------------------
          block call     12.964M (± 0.8%) i/s   (77.14 ns/i) -     66.097M in   5.098814s
   yield with &block     17.186M (± 0.5%) i/s   (58.19 ns/i) -     87.092M in   5.067659s
yield without &block     17.162M (± 0.3%) i/s   (58.27 ns/i) -     86.062M in   5.014579s

Comparison:
   yield with &block: 17186340.4 i/s
yield without &block: 17162477.2 i/s - same-ish: difference falls within error
          block call: 12964086.6 i/s - 1.33x  slower

yield with/without &block is same-ish. For simplicity, use without &block.

Appendix: benchmark script

require 'bundler/inline'
gemfile do
  source 'https://rubygems.org'
  gem 'benchmark-ips'
  gem 'benchmark-memory'
end

def block_call(s, &block)
  block.call(s) if block_given?
end

def yield_with_block(s, &block)
  yield(s) if block_given?
end

def yield_without_block(s)
  yield(s) if block_given?
end

Benchmark.ips do |x|
  x.report("block call") {
    block_call("dummy") do |message|
      message
    end
  }
  x.report("yield with &block") {
    yield_with_block("dummy") do |message|
      message
    end
  }
  x.report("yield without &block") {
    yield_without_block("dummy") do |message|
      message
    end
  }
  x.compare!
end

Docs Changes:

N/A

Release Note:

N/A