Remove fmt: convert I* files to std::format (part 4)

[brianlball]

Purpose

Continues the remove-fmt campaign (#11553, #11588, #11592). Converts the I*
source files from EnergyPlus::format/fmt::format to C++20 std::format and
replaces the custom {:.NR} flag with standard specifiers.

Scope

File	calls converted
InternalHeatGains.cc	148
IceThermalStorage.cc	7
ICEngineElectricGenerator.cc	4

0 {R}/{T} and 0 EnergyPlus::format remain in these files. The DoubleWrapper /
Fortran-format machinery in IOFiles.hh is left untouched (later PR).

Specifier mapping (the bit to sanity-check)

Following the A–H precedent, chosen to minimize regression diffs:

• {:.NR} for |v| >= 0.1 (the common case) -> {:.Nf} — reproduces the old fixed
  output exactly, so zero diff.
• Genuinely-tiny values (CO2 rate ~3.8e-8) -> {:.NE} — matches the old scientific
  mantissa with no truncation; avoids {:.Nf} rounding it to 0.000.
• People counts -> {:.1f} — matches the old 1-decimal echo.

Note on {:#G}/{:G}: this PR doesn't use it, but that's not a departure from
the campaign convention. None of the I* fields are the m3/s/W sizing/flow class
that #11553's {:G} normalization targets — these are density/count echo fields
whose baseline was fixed {:.NR}, so {:.Nf}/{:.NE} reproduce them exactly while
{:#G} would reveal extra sig figs and add avoidable small diffs. (The {:#G} ->
{:.5f} revert in #11592, bd0c1a1048, is the precedent for preferring a fixed
spec where it matches and avoids platform-divergence risk.)

Verification (energyplus-regressions vs develop, local)

• 0 numeric result diffs — ESO/MTR/SSZ/ZSZ/JSON byte-identical.
• Table diffs are display-only (Initialization Summary echo): 503 small + 10 big.
  The 10 big are irreducible x.x5 people-count flips — old {R} rounds half-up,
  std::format rounds half-to-even. Same benign class accepted in Remove fmt - Part 2: Remove {R}, Files A-D #11588/Remove fmt - Part 3: Remove {R}, Files E-H #11592.
• Local regression is Windows-only; CI (Linux/mac) is the authoritative check on
  the specifier choices.

Indentation note

Whitespace-only line changes are clang-format v19 reflow: std::format is 7 chars
shorter than EnergyPlus::format, so wrapped calls collapse under the 150-col
limit. pre-commit run clang-format passes clean.

Question for review

Is the {:.Nf} / {:.NE} / {:.1f} mapping the approach you want for this value
class, and are the 10 half-even count-flip diffs acceptable (consistent with prior
parts)?

[github-actions]

⚠️ Regressions detected on macos-14 for commit 12654dd

Regression Summary

• EIO: 638
• Table Big Diffs: 10
• Table Small Diffs: 533

• View Results
• Download Regressions

[github-actions]

⚠️ Regressions detected on ubuntu-24.04 for commit 12654dd

Regression Summary

• EIO: 642
• Table Big Diffs: 10
• Table Small Diffs: 535

• View Results
• Download Regressions