Eliminate some interface dispatch around object pool use

[geeknoid]

• Use concrete types for fields instead of interface types, which avoids interface dispatch and enables inlining.

The issue or feature being addressed

Details on the issue fix or feature implementation

Confirm the following

• I started this PR by branching from the head of the default branch
• I have targeted the PR to merge into the default branch
• I have included unit tests for the issue/feature
• I have successfully run a local build

[codecov-commenter]

Codecov Report

Merging #1205 (3c63036) into main (71b760a) will decrease coverage by 0.02%.
The diff coverage is 100.00%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more.

@@            Coverage Diff             @@
##             main    #1205      +/-   ##
==========================================
- Coverage   83.31%   83.30%   -0.02%     
==========================================
  Files         277      275       -2     
  Lines        6286     6282       -4     
  Branches     1023     1023              
==========================================
- Hits         5237     5233       -4     
  Misses        844      844              
  Partials      205      205              

Flag	Coverage Δ
linux	?
macos	83.30% <100.00%> (-0.02%)	⬇️
windows	83.30% <100.00%> (-0.02%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...Polly.Core/Hedging/Controller/HedgingController.cs	100.00% <ø> (ø)
src/Polly.Core/Utils/ObjectPool.cs	100.00% <ø> (ø)
...Core/Hedging/Controller/HedgingExecutionContext.cs	100.00% <100.00%> (ø)
...rc/Polly.Core/Utils/CancellationTokenSourcePool.cs	100.00% <100.00%> (ø)

[martintmk]

I am wondering what's the real impact of this change.

Can you run the Polly.Core.Benchmarks (at least for timeout startegy) to see the difference?

edit: the benchmarking needed some love, so I addressed this here:
#1206

[geeknoid]

The benchmark around combining the pool types into one were inconclusive, so I just undid that part of the PR.