docs: add changelogs for v13.4 through v13.11

docs/docs/changelog/v13-11.md

@@ -0,0 +1,13 @@
+---
+slug: "v13.11-release"
+title: "v13.11"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -68
+---
+
+# eCalc v13.11
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Improved validation of pressures set to zero — eCalc now gives a clear error message instead of failing with unhandled errors.

docs/docs/changelog/v13-4.md

@@ -0,0 +1,15 @@
+---
+slug: "v13.4-release"
+title: "v13.4"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -61
+---
+
+# eCalc v13.4
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Fixed a crash that could occur when an upstream choke drops pressure so much that the resulting flow rate exceeds the compressor chart limits.
+- Fixed individual ASV anti-surge control reporting incorrect results due to stale solver state and a faulty success check.
+- Fixed a crash when the inlet rate exceeds the compressor chart's stonewall limit — the solver now correctly reports this as infeasible instead of crashing.

docs/docs/changelog/v13-5.md

@@ -0,0 +1,18 @@
+---
+slug: "v13.5-release"
+title: "v13.5"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -62
+---
+
+# eCalc v13.5
+
+## Changes in results? {/* #changes-in-results */}
+
+> ✔️ **YES**: The bundled NeqSim library has been upgraded from 3.0.38 to 3.8.0. This may cause minor changes in thermodynamic calculations (flash results, fluid properties).
+
+## Features {/* #features */}
+
+- **Pandas 3.x compatibility**: eCalc now supports pandas 3.x in addition to pandas 2.x.
+- Bundled NeqSim upgraded from 3.0.38 to 3.8.0-Java8.

docs/docs/changelog/v13-6.md

@@ -0,0 +1,17 @@
+---
+slug: "v13.6-release"
+title: "v13.6"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -63
+---
+
+# eCalc v13.6
+
+## Features {/* #features */}
+
+- STP: Added `flare` column to STP Export for `FIXED` installations.
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Fixed an issue where the compressor pressure control could incorrectly report a discharge pressure target as unreachable. This occurred when leftover anti-surge recirculation from a previous timestep was carried over, artificially increasing the flow through the compressor and reducing the operating margin available to reach the target pressure.

docs/docs/changelog/v13-7.md

@@ -0,0 +1,21 @@
+---
+slug: "v13.7-release"
+title: "v13.7"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -64
+---
+
+# eCalc v13.7
+
+## Changes in results? {/* #changes-in-results */}
+
+> ✔️ **YES**: Zero-rate compressor timesteps now correctly return 0 instead of NaN. Liquid removal results may also change slightly due to improved NeqSim re-flash handling.
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Compressor timesteps with zero rate now correctly return 0 instead of NaN for rates and power.
+- Fixed incorrect fluid composition downstream of liquid removal. The remaining gas was not re-flashed after liquids were separated out, causing subsequent calculations to use stale fluid properties.
+- Improved robustness of compressor thermodynamic flash calculations near the edge of the operating envelope, reducing NaN propagation and convergence failures at extreme conditions.
+- Fixed mass balance error after liquid removal — the gas stream mass rate was not reduced to account for the separated liquid, overstating the flow rate downstream of the liquid separator.
+- Fixed a potential division-by-zero error when computing maximum standard rate at zero pressure.

docs/docs/changelog/v13-8.md

@@ -0,0 +1,19 @@
+---
+slug: "v13.8-release"
+title: "v13.8"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -65
+---
+
+# eCalc v13.8
+
+## Changes in results? {/* #changes-in-results */}
+
+> ✔️ **YES**: Stream mixing now preserves total enthalpy, giving physically correct outlet temperatures. This may change results for models using stream mixing.
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Fuel must now be defined for the full operational period of a consumer or generator set. Previously, missing fuel definitions for certain periods could silently produce incorrect results. eCalc will now raise a validation error if the fuel definition does not cover the full period. If you encounter this error after upgrading, ensure that your fuel definitions span the required time range.
+- Fixed fuel model not covering the full simulation time range — fuel values could be missing for timesteps outside the fuel definition's own time series.
+- Fixed individual ASV pressure control returning NaN shaft speed instead of a valid result for certain edge-case operating points, which could propagate errors into subsequent calculations.

docs/docs/changelog/v13-9.md

@@ -0,0 +1,13 @@
+---
+slug: "v13.9-release"
+title: "v13.9"
+authors: ecalc-team
+tags: [release, eCalc]
+sidebar_position: -66
+---
+
+# eCalc v13.9
+
+## Bug Fixes {/* #bug-fixes */}
+
+- Fixed a numerical convergence issue where the bisection search could return inaccurate results when intermediate evaluations were rejected.