Simplify apply load limits mode.

[dmkozh]

Description

Now that we have a better understand of what works for the benchmark, we can simplify the config and limit it to just the ledger limits + TPL, and derive the transaction profile automatically (initially we planned a capability of using different tx distributions, but that doesn't seem to be too useful and it does add noise to already noisy benchmarks).

Also remove some unused modes from apply load and loadgen.

Checklist

• Reviewed the contributing document
• Rebased on top of master (no merge commits)
• Ran clang-format v8.0.0 (via make format or the Visual Studio extension)
• Compiles
• Ran all tests
• If change impacts performance, include supporting evidence per the performance document

[Copilot]

Pull request overview

This PR simplifies the apply-load “ledger-limits” benchmarking configuration by removing unused modes and per-transaction distribution knobs, and instead deriving a single apply-load transaction profile directly from the configured ledger/tx limits (plus tx count / TPL).

Changes:

• Remove limits-for-model-tx apply-load mode and the soroban_invoke_apply_load loadgen mode, along with their tests/docs/config parsing.
• Introduce ApplyLoadTxProfile and derive it automatically in ApplyLoad for LIMIT_BASED mode; thread it through to Soroban load transaction generation.
• Update example configs/docs and Windows VS project files to reflect removed/added sources.

Reviewed changes

Copilot reviewed 15 out of 15 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
src/simulation/test/LoadGeneratorTests.cpp	Removes apply-load/realistic-invoke test paths and config knobs that no longer exist.
src/simulation/TxGenerator.h	Adds ApplyLoadTxProfile and updates invoke API to accept it.
src/simulation/TxGenerator.cpp	Uses the derived tx profile for Soroban load V2 generation and adjusts padding/auth sizing logic.
src/simulation/LoadGenerator.h	Removes SOROBAN_INVOKE_APPLY_LOAD mode from the enum.
src/simulation/LoadGenerator.cpp	Removes parsing/execution/status handling for the removed mode.
src/simulation/ApplyLoad.h	Removes “find limits for model tx” declarations; adds accessors and tx-profile state.
src/simulation/ApplyLoad.cpp	Derives ApplyLoadTxProfile in LIMIT_BASED mode and simplifies hot-archive sizing logic.
src/main/Config.h	Removes now-obsolete apply-load config fields and the removed apply-load mode.
src/main/Config.cpp	Removes config parsing for deleted apply-load fields/mode and updates validation message.
docs/software/commands.md	Updates apply-load mode documentation to remove the deleted mode.
docs/apply-load-limits-for-model-tx.cfg	Removes the obsolete example config file.
docs/apply-load-ledger-limits.cfg	Updates example config to match the simplified limits-driven approach.
docs/apply-load-for-meta.cfg	Updates example config to match the simplified limits-driven approach.
Builds/VisualStudio/stellar-core.vcxproj.filters	Updates VS filters to match current ledger/test sources.
Builds/VisualStudio/stellar-core.vcxproj	Updates VS project compilation units/headers accordingly.