`summarize_wqdata()` function throws with some datasets

Two examples using the same EMS_ID/Station but different parameters/variables and in one case the function `summarize_wqdata()` works and provides a summary table while in the other case the table throws an error

Example where it works properly

```
data_works <- tibble::tibble(
EMS_ID = c("0200016", "0200016", "0200016"),
Station = c("ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93","ELK RIVER ABOVE HIGHWAY 93"),
Variable = c("Nitrogen Total", "Nitrogen Total", "Nitrogen Total"),
Code = c("0114", "0114", "0114"),
Value = c(0.844, 0.949, 0.754),
Units = c("mg/L", "mg/L", "mg/L"),
DetectionLimit = c(0.03, 0.03, 0.03),
ResultLetter = c(NA, NA, NA),
Date = c("2021-11-07", "2021-11-21", "2021-12-05"),
Outlier = c(FALSE, FALSE, FALSE),
Site_Renamed = c("ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93","ELK RIVER ABOVE HIGHWAY 93"),
UPPER_DEPTH = as.factor(c(NA, NA, NA)),
Detected = as.factor(c("TRUE", "TRUE", "TRUE")),
Timeframe = as.factor(c("2021", "2021", "2021"))
)

wqbc::summarise_wqdata(
data_works,
by = c("EMS_ID"),
censored = TRUE,
na.rm = TRUE
)
```

Output
```
# A tibble: 1 × 14
Variable EMS_ID n ncen min max mean median lowerQ upperQ sd se lowerCL upperCL

1 Nitrogen Total 0200016 3 0 0.754 0.949 0.849 0.845 0.793 0.901 0.0799 0.0461 0.763 0.944
```

Example where it fails

```
data_fails <- tibble::tibble(
EMS_ID = c("0200016", "0200016", "0200016", "0200016", "0200016"),
Station = c("ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93","ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93"),
Variable = c("Aluminum Total",
"Aluminum Total","Aluminum Total","Aluminum Total","Aluminum Total"),
Code = c("AL-T", "AL-T", "AL-T", "AL-T", "AL-T"),
Value = c(0.031, 0.0192, 0.397, 0.0183, 0.1),
Units = c("mg/L", "mg/L", "mg/L", "mg/L", "mg/L"),
DetectionLimit = c(0.5, 0.5, 0.5, 0.5, 0.5),
ResultLetter = c(NA, NA, NA, NA, NA),
Date = c("2020-01-05","2020-01-27",
"2020-02-02","2020-02-17","2020-03-01"),
Outlier = c(FALSE, FALSE, FALSE, FALSE, FALSE),
Site_Renamed = c("ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93","ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93",
"ELK RIVER ABOVE HIGHWAY 93"),
UPPER_DEPTH = as.factor(c(NA, NA, NA, NA, NA)),
Detected = as.factor(c("FALSE", "FALSE", "FALSE", "FALSE", "FALSE")),
Timeframe = as.factor(c("2020", "2020", "2020", "2020", "2020"))
)

wqbc::summarise_wqdata(
data_fails,
by = c("EMS_ID"),
censored = TRUE,
na.rm = TRUE
)
```

Output
```
Error in names(ret) <- c("mean", "se", LCL(x), UCL(x)) :
'names' attribute [4] must be the same length as the vector [0]
In addition: Warning message:
In survreg.fit(X, Y, weights, offset, init = init, controlvals = control, :
Ran out of iterations and did not converge
Backtrace:
▆
1. └─wqbc::summarise_wqdata(...)
2. └─plyr::ddply(...)
3. └─plyr::ldply(...)
4. └─plyr::llply(...)
5. ├─plyr:::loop_apply(n, do.ply)
6. └─plyr ``(1L)
7. └─wqbc .fun(piece, ...)
8. ├─base::mean(ml)
9. └─NADA::mean(ml)
10. └─NADA .local(x, ...)
```

Start by reproducing the failure with the data_fails example and wqbc::summarise_wqdata(..., by = c("EMS_ID"), censored = TRUE, na.rm = TRUE). Trace the call through plyr::ddply to NADA::mean, where the names error and survreg.fit convergence warning occur. Done means this input returns a summary table without throwing while preserving the working example's behavior.

Written by the indexing model from the issue text.