Tiered charge bug fix and additions

[dalyw]

1. Contributes to resolving flex-pse issue BUG: opex.py never forwards consumption_estimate, so tiered charges silently price at zero flex-pse/flex-pse#73

flex-pse identified EECO bugs/limitation:

• EECO's tiered-charge pricing mutates the charge arrays it is given in place, even though subsequent calculate_cost calls might use different tiers
• Helpful to have warnings for charge tiers that are zeroed

Updates in costs.py:

• Add a ValueError for a consumption_estimate that isn't a scalar or has the wrong length.
• Only apply convex relaxation for tier with a finite next_limit or an hour-specific rate. Allowing flat topmost-tier scenarios to be represented exactly / for an LP without requiring consumption estimate

2. Resolves Demand-charge epigraph: scope to the assessed window, bound _max, expose a handle #60
   With these changes:

utils.py

• index_set, lower_bound, and initialize arguments to ut.max(), passed through to add_component()

costs.py

• adding get_charge_window() helper and using it in get_prev_demand_dict and calculate_demand_cost where it calls ut.max()
• in calculate_demand_cost, calling ut.max() first to get the timesteps that matter for demand charges

tests

• adding test casess to test_max_pyo
• Adding epigraph expectation to test_calculate_cost_pyo_non_standard_index
• New test_get_charge_window.

Code edits and new test cases written with prompting Claude Opus 5 then manually checked

Changes would be a minor version bump, no breaking changes

Pull request recommendations:

• Name your pull request your-development-type/short-description. Ex: feature/gui
• Link to any relevant issue in the PR description. Ex: Resolves [ENH: Support optimization of flexibility metrics #12]
• Provide context of changes.
• Provide relevant tests for your feature or bug fix.
• Provide or update documentation for any feature added by your pull request.

Thanks for contributing!

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.06%. Comparing base (fcbe040) to head (2f10eda).

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #63      +/-   ##
==========================================
+ Coverage   95.21%   96.06%   +0.84%     
==========================================
  Files          10       10              
  Lines        1861     1929      +68     
==========================================
+ Hits         1772     1853      +81     
+ Misses         89       76      -13     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.